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
/
Integrative genomic and epigenomic analysis of human cancer
(USC Thesis Other)
Integrative genomic and epigenomic analysis of human cancer
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
INTEGRATIVE GENOMIC AND EPIGENOMIC ANALYSIS
OF HUMAN CANCER
by
Hui Shen
_______________________________________________
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirement for the Degree
DOCTOR OF PHILOSOPHY
(GENETIC, MOLECULAR AND CELLULAR BIOLOGY)
May 2013
Copyright 2013 Hui Shen
Table of Contents
Dedication iii
Acknowledgements iv
List of Tables vi
List of Figures vii
Abstract x
Chapter One: Introduction and Review 1
Chapter One Bibliography 34
Chapter Two: Serous Ovarian Cancer 55
Chapter Two Bibiliography 86
Chapter Three: Endometrial Cancer 91
Chapter Three Bibiliography 105
Chapter Four: Clear-cell Renal Cell Carcinoma 107
Chapter Four Bibiliography 124
Chapter Five: Epigenetic Analysis Leads to Identification of
HNF1B as a Subtype-Specific Susceptibility
Gene for Ovarian Cancer 127
Chapter Five Bibiliography 156
Conclusion 161
Bibliography 165
Appendices
Appendix A: A List Of the 168 Candidate Epigenetically Silenced
Genes in serous ovarian cancer 195
Appendix B: OCAC Study Sites 199
Dedication
To
My parents,
For never doubting their daughter,
And making all of this possible.
Acknowledgements
My first debt of deepest gratitude must go to my advisor, Dr. Peter W. Laird. I
consider myself extremely lucky to have Dr. Laird as my mentor - one cannot ask for a
better advisor than a scientist who not only excels in his field but is also of great integrity.
I am grateful for all the knowledge that he generously imparted, the unwavering support
and trust he never ceased to offer, the great opportunities that he gave me, but even
more, for the great impact that he has on me as a scientist, and also as a person.
I would also like to thank my dissertation committee members Dr. Leigh Pearce
and Dr. Benjamin Berman, who are not only great teachers but also cherished friends. I
would especially like to thank Dr. Leigh Pearce, without whose great insight and passion
for science the last chapter of this dissertation would not have been possible.
I am obliged to many of my colleagues and friends in USC who have supported
me and offered tremendous help in writing this dissertation, especially Daniel
Weisenberger, Timothy Triche Jr, Toshi Hinoue, Houtan Noushmehr, Moiz Bootwalla,
Dennis Maglinte, Douglas Stram, Simeen Malik, Swapna Mahurkar, Janice Galler,
Yaping Liu, Lijing Yao, Zack Ramjan and Vasu Punj from the USC Epigenome Center,
Mihaela Campan, Shirley Oghamian and Kwangho Lee from the Laird Lab, Alice Lee,
Kate Lawrenson, Dr. Simon Gayther, Suhn Rhie, Sheng-Fang Su, Wan-Ting Chen and
Fan Ding.
This dissertation would also not have been possible without the numerous
collaborators in the Cancer Genome Atlas (TCGA) and Ovarian Cancer Association
Consortium (OCAC). I will not list every one of them here, but they have contributed to
this dissertation as well and I owe them my sincere thanks.
I would also like to thank USC Provost’s PhD Fellowship, Charles Heidelberger
Memorial Predoctoral Scholarship, and NIH Training Grant T32GM067587 for financial
support in writing this dissertation.
List of Tables
Table 2-1: Distribution of BRCA inactivation events across promoter
methylation clusters 71
Table 2-2: Overlap between gene expression and DNA methylation
Subtypes 72
Table 3-1: MLH1 epigenetic silencing differs by histological subtype,
and is strongly associated with microsatellite instability (MSI)
93
Table 3-2: Distributation of samples by analytical batch and platform
99
Table 5-1: Association between invasive, serous and clear cell ovarian
cancer for ten HNF1B SNPs that reached genome-wide
significance in Whites 132
List of Figures
Figure 1-1: Representative Epigenetic States 3
Figure 1-2: Histone H3 Lysine Writers, Erasers and Readers 6
Figure 1-3: Genetic Alterations in Epigenetic Regulators 20
Figure 1-4: Genetic Disruption of Epigenetic Control at H3K27 in
Cancer 32
Figure 2-1: BRCA1 gene expression versus promoter methylation 60
Figure 2-2: BRCA1 epigenetic silencing validation 62
Figure 2-3: ROC curve of BRCA1 methylation validation by
MethyLight 63
Figure 2-4: RAB25 gene expression versus promoter methylation 64
Figure 2-5: AMT gene expression versus promoter methylation 65
Figure 2-6: SPARCL1 gene expression versus promoter
methylation 66
Figure 2-7: CCL21 gene expression versus promoter methylation 67
Figure 2-8: DNA methylation subtypes 69
Figure 2-9: DNA methylation subtypes- Clinical implications 70
Figure 3-1: MLH1 is methylated across the CpG island associated with
its promoter in about 30% of the tumor samples 92
Figure 3-2: Unsupervised clustering of the DNA methylation data reveals
four subtypes 94
Figure 3-3: The serous-like tumors (MC3) have extensive copy-number
changes 95
Figure 3-4: The serous-like tumors also exhibit loss of DNA methylation
at X-linked loci 96
Figure 3-5: Similarity between the serous-like endometrial tumors, serous
ovarian tumors, and basal-like breast tumors 97
Figure 3-6: Cross-tumor comparison 98
Figure 3-7: HM450 versus HM27, before and after platform
Correction 103
Figure 4-1: Unsupervised clustering of DNA methylation data for
373 tumors 109
Figure 4-2: Promoter DNA hypermethylation increases with tumor
stage/grade 110
Figure 4-3: 290 epigenetically silenced genes in ccRCCs 111
Figure 4-4: VHL epigenetic silencing 112
Figure 4-5: UQCRH epigenetic silencing 113
Figure 4-6: RASSF1A is not epigenetically silenced in ccRCCs
compared to the normals 114
Figure 4-7: A distinct DNA methylation profile associated with
SETD2 mutation 115
Figure 4-8: Permutation of the SETD2 label 116
Figure 5-1: Identification of HNF1B as a subtype-specific candidate
gene for ovarian cancer and establishment of it as a
susceptibility gene 130
Figure 5-2: Genetic variants in the HNF1B locus are associated with
risk of ovarian cancer histological subtypes 131
Figure 5-3: HNF1B promoter DNA methylation, protein expression and
global DNA methylation pattern by subtype 133
Figure 5-4: HNF1B promoter methylation is unlikely to be a passenger
event by global DNA methylation changes 134
Figure 5-5: Correlation of serous risk-associated SNPs with HNF1B
promoter DNA methylation level 136
Figure 5-6: Validation of the SNP-DNA methylation association with
TCGA data 137
Figure 5-7: HNF1B DNA methylation levels across the entire promoter
region differ by rs11658063 genotype 138
Figure 5-8: Phenotypic effects and downstream targets of HNF1B
overexpression in immortalized endometriosis cells 139
Figure 5-9: Linkage disequilibrium plot of the genome-wide significant
serous and clear cell SNPs as well as the SNPs associated
with prostate and uterine cancer and diabetes 142
Abstract
Cancer arises as a consequence of cumulative disruptions to cellular growth control,
with Darwinian selection for those heritable changes which provide the greatest clonal
advantage. These traits can be acquired and stably maintained by either genetic or
epigenetic means. Alterations in the genome and epigenome could influence each other
and cooperate to promote oncogenic transformation. Disruption of epigenomic control is
pervasive in malignancy, and can be classified as an enabling characteristic of cancer
cells, akin to genome instability and mutation. We examined epigenetic profiles of
several human cancers, including ovarian, endometrial and clear cell renal cell
carcinoma (ccRCC) in the context of other genomic alterations, as part of the Cancer
Genome Atlas (TCGA) project. We found varying degrees of disease heterogeneity
among tumors of the various cancer types studied. We found that endometrial cancer
comprises several distinct molecular groups, with a serous-like subtype that is similar to
serous ovarian cancer and basal-like breast cancer. We also identified important or
potentially important epigenetically silenced genes and studied their clinical implications.
BRCA1 is epigenetically silenced in 12% of serous ovarian cancer cases, and mutually
exclusive with BRCA1/2 mutations. This epigenetic silencing is associated with worse
prognosis. VHL epigenetic silencing in ccRCC is mutually exclusive with VHL mutation.
Finally, we hypothesized that genetic variants in one gene that we found to be
epigenetically silenced in serous ovarian cancer and differentially methylated among
different ovarian cancer subtypes, HNF1B, would be associated with ovarian cancer risk
in a subtype-specific way. We comprehensively mapped variation in HNF1B with respect
to EOC risk. Different SNPs were associated with invasive serous (rs7405776 OR=1.13,
p=3.1x10
-10
) and clear cell (rs11651755 OR=0.77, p=1.6x10
-8
) ovarian cancers. Risk
alleles for the serous subtype were associated with higher HNF1B promoter methylation
in these tumors. Unmethylated, expressed HNF1B, primarily present in clear cell tumors,
coincided with a CpG Island Methylator Phenotype (CIMP) affecting numerous other
promoters throughout the genome. Different variants in HNF1B are associated with risk
of serous and clear cell ovarian cancers; DNA methylation and expression patterns are
also notably distinct between these subtypes. These findings underscore distinct
mechanisms driving different ovarian cancer histological subtypes.
1
Chapter One: Introduction and Review
Introduction
Cancer develops through successive disruptions to the controls of cellular
proliferation, immortality, angiogenesis, cell death, invasion and metastasis. This
evolutionary process requires newly acquired malignant traits to be stably
encoded so that oncogenic events can accumulate in clonal lineages. Genetic
mechanisms of mutation, copy number alteration, insertion, deletion and
recombination are particularly well suited as vehicles of persistent phenotypic
change. For this reason, cancer has long been viewed as a disease based
principally on genetics. Nevertheless, genetic events occur at low frequency, and
are thus not a particularly efficient means for malignant transformation. Some
cancer cells overcome this bottleneck by acquiring DNA repair defects, thus
boosting the mutation rate. Mechanisms of epigenetic control offer an alternative
path to acquiring stable oncogenic traits. Epigenetic states are flexible yet persist
through multiple cell divisions, and exert powerful effects on cellular phenotype.
Although cancer cells have long been known to undergo epigenetic changes,
genome-scale genomic and epigenomic analyses have only recently revealed
the widespread occurrence of mutations in epigenetic regulators and the breadth
of alterations to the epigenome in cancer cells (You and Jones, 2012). It is now
clear that genetic and epigenetic mechanisms influence each other, and work
cooperatively to enable the acquisition of the hallmarks of cancer (Hanahan and
Weinberg, 2011).
2
Shaping the Epigenome
Epigenetic mechanisms allow genetically identical cells to achieve diverse
stable phenotypes by controlling the transcriptional availability of various parts of
the genome through differential chromatin marking and packaging. These
embellishments include direct DNA modifications, primarily CpG cytosine-5
methylation (Jones, 2012), but also hydroxylation, formylation and carboxylation
(Ito et al., 2011), as well as nucleosome occupancy and positioning (Gaffney et
al., 2012; Valouev et al., 2011), histone variants, and dozens of different histone
modifications (Tan et al., 2011b), interacting proteins (Ram et al., 2011), and
non-coding RNAs (Fabbri and Calin, 2010; Lee, 2012). These epigenetic marks
do not act in isolation, but form a network of mutually reinforcing or counteracting
signals. Genome-scale projects charting the human epigenome are rapidly
extending our understanding of epigenetic marks and how they interact (Adams
et al., 2012; Encode-Project-Consortium et al., 2012; Ernst et al., 2011).
A key facet of epigenetics is that these marks can be stably maintained,
yet adapt to changing developmental or environmental needs. This delicate task
is accomplished by initiators, such as long non-coding RNAs, writers, which
establish the epigenetic marks, readers, which interpret the epigenetic marks,
erasers, which remove the epigenetic marks, remodelers, which can reposition
nucleosomes, and insulators, which form boundaries between epigenetic
domains. Epigenetic writers are directed to their target locations by sequence
context, existing chromatin marks and bound proteins, non-coding RNAs, and/or
nuclear architecture. Those marks are then recognized by reader proteins to
convey information for various cellular functions. The establishment,
maintenance, and change of epigenetic marks are intricately regulated, with
3
crosstalk among the marks and writers to help guide changes to the epigenetic
landscape.
DNA Methylation
De novo methylation of DNA is catalyzed by the enzymes DNMT3A and
DNMT3B, and is then maintained by the major DNA methyltransferase DNMT1,
with participation from DNMT3A and DNMT3B (Jones and Liang, 2009). DNA
methylation patterns are guided in part by primary DNA sequence context (Cedar
and Bergman, 2012; Lienert et al., 2011) and influenced by germline variation
(Gertz et al., 2011; Kerkel et al., 2008). Much of the mammalian genome consists
Figure 1-1: Representative Epigenetic States. Examples of representative
epigenetic states are shown for several typical categories of genes and in different
cellular contexts. A. CpG-poor promoters are often tissue-specific and/or reside in
inducible genes which can be readily turned on or off. B. Transcription factor (TF)
binding at regulatory elements and the promoter initiates nucleosome depleted
regions (NDR). Genes with CpG island promoters are often constitutively expressed
housekeeping genes or are repressed by the Polycomb complexes, such as
transcription factor master regulators of differentiation and development in stem cells.
C. Polycomb targets in stem cells are predisposed to cancer-specific
hypermethylation.
4
of vast oceans of DNA sequence containing sparsely distributed, but heavily
methylated CpG dinucleotides, punctuated by short regions with unmethylated
CpGs occurring at higher density, forming distinct islands in the genome (Bird et
al., 1985). These CpG islands (CGIs) are protected from DNA methylation in part
by GC strand asymmetry and accompanying R-Loop formation (Ginno et al.,
2012) and possibly also by active demethylation mediated by the TET family
members (Williams et al., 2012). The unmethylated state of CpG islands in the
germline, along with biased gene conversion, helps to preserve CpG islands,
despite ongoing attrition of methylated CpG dinucleotides by cytosine
deamination throughout most of the genome (Cohen et al., 2011). Transition
zones between CpG islands and CpG oceans are called CpG shores, and
display more tissue-specific variation in DNA methylation (Irizarry et al., 2009).
CpG islands span the transcription start sites of about half of the genes in the
human genome, largely representing genes that are either actively expressed or
poised for transcription (Figure 1-1).
Methylated DNA is recognized by methyl-CpG binding domains (MBD) or
C2H2 zinc fingers. The MBD-containing DNA methylation readers include MBD1,
MBD2, MBD4 and MeCP2, whereas Kaiso (ZBTB33), ZBTB4 and ZBTB38
proteins use zinc fingers to bind methylated DNA. MBDs and Kaiso are believed
to participate in DNA methylation mediated transcriptional repression of tumor
suppressor genes with promoter DNA methylation.
Histone Modifications
Post-translational modifications of histones are coordinated by
counteracting histone-methyltransferases (HMTs) and demethylases (e.g. KDMs),
histone acetyltransferases (HATs) and deacetylases (HDACs), and writers and
erasers of phosphorylation, as well as many other modifications (Chi et al., 2010;
5
Tan et al., 2011b). These histone modifiers generally act in complexes, such as
the repressive Polycomb (PcG) and activating Trithorax (TrxG) group complexes,
which counterbalance each other in the regulation of genes important for
development, but which have also been implicated in cancer (Mills, 2010).
Polycomb repressive complexes (PRCs) are guided to their targets in part by
intrinsic signals in the genome sequence (Ku et al., 2008; Tanay et al., 2007).
The histone H3K27me3 mark deposited by PRC2 provides docking sites for
PRC1, whose enzymatic core unit RING1B monoubiquitinylates histone H2A at
lysine 119 (H2AK119ub1) thereby blocking RNA polymerase II elongation. The
Trithorax group complex, containing MLL, which lays down the H3K4 methylation
mark, counteracts Polycomb function. The transcription factors encoding master
regulators of differentiation and development are targeted by PRC2 in embryonic
stem cells and held in a bivalent chromatin state poised for transcription, with
both the activating H3K4me3 and the repressive H3K27me3 (Bernstein et al.,
2006) (Figure 1-1). During differention the Trithorax demethylase, KDM6A/UTX
removes the repressive H3K27me3 mark, allowing transcription elongation to
proceed for genes required in that particular lineage, while genes not required in
that cell type undergo expansion of the H3K27me3 mark, and may acquire
H3K9me3 additionally (Hawkins et al., 2010), which is recognized by readers like
HP1 to reinforce a repressive state. Other histone marks have various readers
with binding motifs including bromodomain, PHD domain, chromodomain, and
tudor domain (Musselman et al., 2012) (Figure 1-2). Trithorax and Polycomb
complexes recruit HATs and HDACs, respectively, to counteract each other, and
the establishment of histone acetylation can block Polycomb binding (Mills, 2010).
6
Histone Variants
Histone variants provide an additional layer of regulation. The main
histone genes have multiple copies in the genome and are expressed during S-
phase. Single-copy variants are also expressed at other phases of the cell cycle
and have distinct functions and/or locations. H2A has the largest number of
variants, including H2A.Z, MacroH2A, H2A-Bbd, H2AvD, and H2A.X (Kamakaka
Figure 1-2: Histone H3 Lysine Writers, Erasers and Readers. Although many
other important histone modifications also occur, only major histone H3 lysine
modifications (Ac: Acetylation; me1: monomethylation; me3: trimethylation) with
well-defined functions are shown above a representative gene, with red and green
respectively indicating general association with transcriptional repression and
activation. Epigenetic regulators are listed to the right of each mark. Acetylation
across different lysines share writers and erasers, while methylation usually has
dedicated enzymes. Readers (which can also be writers and erasers themselves)
recognize different chromatin states and propagate the signal in various ways,
including self reinforcement or cross talk, transcriptional activation or repression, or
DNA repair. Crosstalk can also occur between histone modification and DNA
methylation, since DNMT3A, DNMT3L, UHRF1 all contain reader domains for
chromatin states.! Superscripts to the ‘readers’ column indicate the reader domain
type: 1 – PHD; 2 – Chromo Domain; 3 – ADD (ATRX-DNMT3-DNMT3L); 4 – Tudor;
5 – Ankyrin Repeats; 6 – Bromodomain; 7 – WD40; 8- PWWP.
7
and Biggins, 2005). The H3 variants include H3.3 and centromeric H3 (CenH3, or
CENP-A), as well as a mammalian testis-specific histone H3 variant called H3.4.
Nucleosomes containing H3.3 and H2A.Z are located at dynamic regions
requiring nucleosome mobility and exchange, such as at actively expressed gene
promoters (Jin et al., 2009). Wide presence of H2A.Z in embryonic stem cells
(Zhu et al., 2013) suggests prevalent chromatin exchange, consistent with the
emerging idea that the genome of ESC is generally kept highly accessible.
During differentiation H2A.Z quickly redistributes. The mechanisms of recruitment
have not been fully delineated, but various chromatin remodeler complexes
and/or chaperones have been shown to be involved. For example, SRCAP is
involved in H2A.Z loading into promoter/TSS, while H3.3 is loaded to
telomeric/pericentric regions by the ATRX/DAXX complex and promoter/TSS by
HIRA (Boyarchuk et al., 2011).
Nucleosome Positioning and Remodeling
The positioning of nucleosomes displays a weak 10-bp periodicity
associated with minor sequence composition fluctuations in phase with the DNA
helical repeat. Some nucleosomes are more consistently positioned in phased
arrays anchored by sequence-specific binding of proteins such as CTCF or
adjacent to nucleosome-free regions at transcription start sites (Gaffney et al.,
2012; Valouev et al., 2011). CpG islands have been associated with
transcription-independent nucleosome-depletion at mammalian promoters
(Fenouil et al., 2012). ATP-dependent nucleosome remodeling complexes are
responsible for sliding of the nucleosomes, as well as insertion and ejection of
histone octamers, processes important for transcriptional repression and
activation, and other important cellular functions such as DNA replication and
repair. The remodeling complexes can be divided into four families: SWI/SNF,
8
CHD (chromodomain and helicase-like domain), ISWI, and INO80 (including
SWR1, or SRCAP in mammals).
Insulators
The CCCTC-binding factor CTCF (and its paralogue CTCFL/BORIS,
expressed in the germline) are the only insulator proteins that have been
identified so far in vertebrates. CTCF has a strong binding motif and there is
extensive overlap of the occupied CTCF-binding sites among different cell types
(Kim et al., 2007). CTCF binds to enhancer blocking elements to prevent
enhancer interactions with unintended promoters (‘enhancer blocking insulator’),
and also demarcates active and repressive chromatin domains (‘barrier
insulator’).
Nuclear Architecture
The genome can be compartmentalized based on nuclear architecture
and associated genomic features into mostly heterochromatic late-replicating
regions attached to the nuclear lamina at the nuclear periphery, and more gene-
rich early-replicating regions closer to the nuclear interior (Encode-Project-
Consortium et al., 2012; Meuleman et al., 2012). Lamina-associated sequences
(LASs) enriched for a GAGA motif are bound by transcriptional repressors, and
appear to contribute to the establishment of lamina associated domains (LADs)
in the mammalian genome (Zullo et al., 2012).
Maintaining the Epigenetic State
The persistence of epigenetic traits in a growing tumor requires that the
epigenome be faithfully copied during cell division. The chromatin structure is
dismantled for passage of the replication fork (RF). Newly synthesized DNA and
histone octamers are then assembled at the RF by Chromatin assembly factor I
(CAF1), tethered to the RF by PCNA. Similarly, the dedicated maintenance DNA
9
methyltransferase DNMT1, the euchromatic H3K9 methyltransferase G9a,
among other epigenetic maintainers, are loaded to RFs and copy the epigenetic
marks. The Trithorax and Polycomb complexes are recruited prior to replication
and distributed evenly to the mother and daughter strands at the RF, and restore
the correct marks on the daughter molecules during G1 (Petruk et al., 2012). The
histone marks are self-reinforcing and self-propagating, as PcG, SUV39H1/2,
SETDB1 and TrxG all bind to the marks that they are responsible for catalyzing,
via an intrinsic reader domain or by interacting with a reader protein, thus helping
to maintain the epigenetic state. Nucleosomes containing methylated DNA also
stabilize DNMT3A/3B, which is a self-reinforcing mechanism for DNA methylation
maintenance (Sharma et al., 2011).
10
Disruption of Epigenetic Control in Cancer
Most studies of cancer epigenetics have focused on DNA methylation, as
the epigenetic mark that most easily survives various forms of sample processing,
including DNA extraction, and even formalin fixation and paraffin embedding
(Laird, 2010). However, other epigenetic marks also undergo broad changes,
including long non-coding RNAs and miRNAs (Baer et al., 2013; Baylin and
Jones, 2011; Dawson and Kouzarides, 2012; Sandoval and Esteller, 2012), and
loss of K16 acetylation and K20 trimethylation at histone H4 (Fraga et al., 2005;
Hon et al., 2012; Kondo et al., 2008; Seligson et al., 2005; Yamazaki et al., 2013).
Loss of 5-methylcytosine in cancer cells was discussed more than three decades
ago (Ehrlich and Wang, 1981), with global DNA hypomethylation reported in
cancer cell lines (Diala and Hoffman, 1982; Ehrlich et al., 1982) and reduced
levels of DNA methylation found at selected genes in primary human tumors
compared to normal tissues (Feinberg and Vogelstein, 1983). The widespread
loss of DNA methylation contrasted starkly with the subsequent finding of
hypermethylation of CpG islands in cancer (Baylin et al., 1986), including of
promoter CpG islands of tumor-suppressor genes (Jones and Baylin, 2002).
These seemingly contradictory findings have been widely reported for many
types of cancer (Baylin and Jones, 2011).
The causal relevance of epigenetic changes in cancer was initially
questioned but this concern has now largely been laid to rest. First, many known
tumor-suppressor genes have been shown to be silenced by promoter CpG
island hypermethylation (Jones and Baylin, 2002). Importantly, the finding that
these silencing events are mutually exclusive with structural or mutational
inactivation of the same gene, such as the case for BRCA1 in ovarian cancer
(TCGA, 2011) and for CDKN2A in squamous cell lung cancer (TCGA, 2012a),
11
reinforces the concept that epigenetic silencing can serve as an alternative
mechanism in Knudson's two-hit hypothesis (Jones and Laird, 1999). Second,
mouse models of cancer have been shown to require epigenetic writers and
readers for tumor development (Laird et al., 1995; Prokhortchouk et al., 2006;
Sansom et al., 2003). Third, some DNA methylation changes appear to be
essential for cancer cell survival, suggesting an acquired addiction to epigenetic
alterations (De Carvalho et al., 2012). Finally, a plethora of significantly mutated
epigenetic regulators have now been reported for many types of human cancer,
as discussed further below.
Long-Range Coordinated Disruptions and Nuclear Architecture
The genome of undifferentiated embryonic stem cells is uniformly heavily
methylated across CpG oceans, punctuated by unmethylated CpG islands. As
stem cells differentiate and proliferate, the late-replicating lamin-associated
domains (LADs) undergo progressive loss of DNA methylation within CpG
oceans, and the LADs become recognizable as long partially methylated
domains (PMDs), which become even more strikingly demarcated as
hypomethylated domains in cancer cells (Berman et al., 2012; Hansen et al.,
2011; Hon et al., 2012; Lister et al., 2009). This loss of DNA methylation is
associated with an increase of repressive chromatin with large organized
chromatin-lysine-(K) modification regions (LOCKs) (Hansen et al., 2011; Hon et
al., 2012; Lister et al., 2009). CpG island hypermethylation is enriched in the
hypomethylated domains, suggesting that these two events may be
mechanistically linked, but confined to distinct areas of the genome near the
nuclear periphery (Berman et al., 2012). These long regions of DNA
hypomethylation and repressive chromatin are consistent with prior reports of
coordinated epigenetic silencing events located across megabase distances, a
12
phenomenon termed Long-Range Epigenetic Silencing (LRES) (Clark, 2007;
Coolen et al., 2010).
It is noteworthy that the euchromatic part of the genome associated with
the interior of the nucleus is generally much more epigenetically stable during cell
differentiation, aging and malignant transformation. However, loss of the DNA
methyltransferase Dnmt3a can promote tumor progression with uniform
hypomethylation across the genome, and moderate deregulation of genes in
euchromatic regions (Raddatz et al., 2012).
Disruption of Differentiation and Development
Differences between cell types are guided by the expression of tissue-
specific transcription factors and consolidation of associated epigenetic states.
Therefore, the epigenome of a cancer cell is determined in part by the cell of
origin for that cancer and includes passenger hypermethylation events at genes
not required in that particular lineage (Sproul et al., 2012). Epithelial to
mesenchymal transition (EMT) of cancer cells is partly under reversible
epigenetic control (Craene and Berx, 2012). For example, primary breast tumors
display heterogeneous and unstable silencing of the CDH1 (E-cadherin) gene,
which facilitates the plasticity required during extravasation, metastasis and
establishment of a solid tumor at the metastatic site (Graff et al., 2000).
It has long been debated whether cancer cells arise by dedifferentiation
or instead originate from stem cells or early progenitors by a differentiation block.
Polycomb repressors mark genes in stem cells encoding master regulators of
differentiation and development, poised to either be turned on to coordinate
differentiation of a lineage, or to be fully repressed if it is not needed in that
particular lineage (Bernstein et al., 2006). These genes occupied by Polycomb
repressors in stem cells are particularly prone to acquiring CpG island
13
hypermethylation during cell proliferation, aging and particularly malignant
transformation (Ohm et al., 2007; Schlesinger et al., 2007; Teschendorff et al.,
2010; Widschwendter et al., 2007) (Figure 1-1). Although the genes affected by
this process are primarily those not required or expressed in that particular cell
lineage, cancer cells do also show evidence of silencing of genes essential for
differentiation of their cell of origin (Berman et al., 2012; Easwaran et al., 2012;
Gal-Yam et al., 2008; Mohn et al., 2008; Teschendorff et al., 2010). This
predisposition of Polycomb target genes to aberrant permanent epigenetic
silencing is consistent with a model in which stem cells slowly acquire irreversible
silencing of poised master regulators required for successful differentiation. As a
consequence, some stem cells lose their ability to properly differentiate while
retaining their self-renewal capabilities, and become attractive candidates for
malignant transformation by subsequent genetic and epigenetic events. One
provocative implication of this model is that the first steps of oncogenesis may in
some cases be an epigenetic defect affecting the differentiation capabilities of
stem cells, as opposed to a gatekeeper mutation.
Hematopoietic cell lineages and their corresponding malignancies also
offer insights into the role of epigenetics in differentiation and transformation. For
example, the DNMT3A gene is commonly mutated in human cases of Acute
Myeloid Leukemia (AML) (Ley et al., 2010; Yan et al., 2011), whereas loss of
Dnmt3a in mice progressively impairs hematopoietic stem cell differentiation
(Challen et al., 2012), suggesting that epigenetic perturbation can lead to
differentiation block and subsequent malignant transformation.
CpG Island Methylator Phenotypes
Aberrant DNA methylation of promoter CpG islands in cancer was initially
viewed as a spontaneous or stochastic event with selection for functionally
14
relevant silencing events. However, the discovery of cases of colorectal cancer
with an exceptionally high frequency of CpG island hypermethylation suggested
a coordinated event, possibly attributable to an epigenetic control defect. This
phenomenon was referred to as a "CpG Island Methylator Phenotype" (CIMP)
(Toyota et al., 1999), analogous to the mutator phenotypes observed in
mismatch repair deficient cancers. Although the existence of CIMP subsets of
cancer was initially disputed (Yamashita et al., 2003), more recent genome-scale
analyses have unambiguously documented distinct epigenetic subtypes for some
types of cancer, such as colorectal cancer (Hinoue et al., 2012; TCGA, 2012b)
and glioblastoma (Noushmehr et al., 2010), and not for others, such as serous
ovarian cancer (TCGA, 2011). The most distinct examples of CIMP show
exceptionally strong associations with other molecular or pathological features of
the tumors, lending further validity to the biological relevance to this classification.
For example, colorectal CIMP is very tightly associated with the V600E mutation
of the BRAF oncogene (Weisenberger et al., 2006), while glioma CIMP (G-CIMP)
is exceptionally tightly associated with mutation of the IDH1 gene (Noushmehr et
al., 2010). In the case of G-CIMP, IDH1 mutation appears to be a causal
contributor to the phenotype (Turcan et al., 2012), whereas BRAF mutation does
not appear to be directly implicated in colorectal CIMP (Hinoue et al., 2009). The
affected gene subsets,differ between colorectal CIMP and glioblastoma G-CIMP,
and their predisposition to aberrant methylation appears to be distinct from the
susceptibility of stem cell polycomb targets in lamin-attachment domains (Hinoue
et al., 2012), which is generally not restricted to cancer subtypes. Despite a clear
rationale for the association of IDH1 mutation with G-CIMP, the mechanistic
basis for the coordinated hypermethylation events in most cases of CIMP is
unknown, and will remain an active area of investigation.
15
Epigenetic Influences on Genomic Integrity
Mutation rates vary strikingly across the genome, with strong local
influences of base composition on single nucleotide variation (SNV), and regional
effects of sequence composition, chromatin structure, replication timing,
transcription and nuclear architecture, among others on both SNVs and structural
alterations (Hodgkinson and Eyre-Walker, 2011). Despite widespread misuse of
the term in the literature, it should be recognized that mutation rates of a tumor
cannot be inferred directly from observed mutation numbers or frequencies in a
tumor without consideration of the number of cell divisions that have occurred
since a shared reference genome, although comparisons across the genome
obviate the need for Luria-Delbrück fluctuation modeling and analysis. Epigenetic
mechanisms can influence both the rates at which lesions arise and the rates at
which they are repaired. For example, the epigenetic mark 5-methylcytosine
undergoes spontaneous deamination at higher rates than unmethylated
cytosines (Wang et al., 1982), while epigenetic silencing of the MLH1 mismatch
repair gene increases mutation frequencies by several orders of magnitude,
providing an adaptive advantage to mismatch repair deficient cancer cells.
Unmethylated and methylated cytosine residues both undergo
spontaneous hydrolytic deamination but yield uracil and thymine respectively.
Uracil is not a normal constituent base in DNA, and is repaired much more
efficiently than thymine in a mismatch with guanine. As a consequence, the rate
of C-to-T mutations in the context of CpG dinucleotides, most of which contain
methylated cytosines, is about ten-fold higher than any other SNV in the human
genome (Hodgkinson and Eyre-Walker, 2011). This effect is particularly
pronounced in highly proliferative tissues because deamination of 5-
methylcytosine in the parent strand just prior to DNA replication results in a full
16
T:A base substitution that is not recognizable as a lesion for repair.
Approximately a quarter of all TP53 mutations in human cancer are thus
attributable to this epigenetic mark (Olivier et al., 2010).
Regional Effects of Chromatin Organization
Chromatin regulators play a role in maintaining genomic integrity
(Papamichos-Chronakis and Peterson, 2012) and regional chromatin structure
has a major impact on mutation frequencies. Megabase regions of repressive
chromatin, represented by the H3K9me3 mark are positively correlated with
single nucleotide variations in cancer (Schuster-Bockler and Lehner, 2012), while
open chromatin associated with DNAse I hypersensitive sites (DHS) have a
lower inferred mutation rate, but this is partly due to evolutionary constraints on
this compartment (Hodgkinson and Eyre-Walker, 2011). Transcription-coupled
repair may also play a role in suppressing observed mutation frequencies in
gene-rich euchromatic regions.
Other types of mutation and structural change also appear to be
associated with chromatin states. For example, retrotransposition occurs more
frequently in hypomethylated regions (Lee et al., 2012). Genes resistant to
cancer-associated hypermethylation are more likely to have SINE and LINE
retrotransposons near their transcription start sites than methylation-prone genes
(Estecio et al., 2010). Severe hypomethylation appears to be associated with
genomic instability. Mouse models of DNA methyltransferase deficiency display
chromosomal instability (Eden et al., 2003), and germline mutations of the
DNMT3B gene cause ICF syndrome, characterized by centromeric instability
(Okano et al., 1999). Indeed, areas of hypomethylation in the human germline
showed higher frequencies of structural mutability (Li et al., 2012). DNA
17
breakpoints associated with somatic copy-number alterations are also enriched
in hypomethylated domains (De and Michor, 2011).
Epigenetic Influences on DNA Repair
Depletion of DNA methyltransferases causes increased microsatellite
instability (Guo et al., 2004; Kim et al., 2004), destabilization of repeats (Dion et
al., 2008), and dramatically increased telomere length, telomeric recombination,
and alternative telomere lengthening (Gonzalo et al., 2006). These effects of
DNA methyltransferase depletion appear to be mediated in part by a drop in DNA
repair proteins as part of DNA damage response (Loughery et al., 2011). The
Dnmt1 protein has also been shown to be recruited to areas of irradiation-
induced DNA damage, possibly to facilitate repair of epigenetic information
following DNA repair (Mortusewicz et al., 2005). It is increasingly appreciated that
chromatin can serve as a cellular sensor for DNA damage and other genomic
events (Johnson and Dent, 2013).
Epigenetic silencing of DNA repair genes such as MLH1, MGMT, BRCA1,
WRN, FANCF, and CHFR can boost mutation rates and promote genomic
instability in cancer cells (Toyota and Suzuki, 2010). Familial cases of tumors
with microsatellite instability (MSI) in Lynch syndrome result from germline
mutations in mismatch repair genes, primarily MSH2 and MLH1. However, most
MSI-high tumors arise from an epigenetic defect in sporadic cases of cancer.
Approximately 15% of sporadic cases of colorectal cancer display MSI as a
consequence of epigenetic silencing of the MLH1 mismatch repair gene by
promoter CpG island hypermethylation (Herman et al., 1998) in the context of
CIMP (Toyota et al., 1999; Weisenberger et al., 2006). MSI caused by epigenetic
silencing of MLH1 has also been reported in other types of cancer, including
about a quarter of sporadic endometrial cancers, (Simpkins et al., 1999).
18
Germline variants of MLH1 and MSH2 can predispose to extensive somatic
epigenetic silencing of these genes, and thereby increase cancer risk (Hitchins et
al., 2011; Ligtenberg et al., 2009). Such familial cases of systemic epigenetic
abnormalities can masquerade as germline transmission of epigenetic defects.
True transgenerational epigenetic inheritance is evident in genomic imprinting,
and in mouse models, but has been difficult to demonstrate directly in human
populations, although there is indirect evidence for its existence (Daxinger and
Whitelaw, 2012).
The O6-Methylguanine DNA methyltransferase (MGMT) enzyme repairs
O6-alkylated guanine residues in genomic DNA. O6-methylguanine pairs with
thymine, and would lead to a G to A transition during DNA replication if left
unrepaired. MGMT promoter methylation in colorectal cancer is associated with
G-to-A mutations in KRAS (Esteller et al., 2000b) and in TP53 (Esteller et al.,
2001). Alkylating agents such as Temozolomide are current standard of care for
malignant glioblastoma (GBM), but are counteracted by MGMT-mediated repair
of the alkylation damage. Epigenetic silencing of MGMT by promoter CpG island
hypermethylation inactivates this repair pathway and renders the tumor more
sensitive to the Temozolomide treatment (Esteller et al., 2000a; Hegi et al., 2005).
19
A Genetic Basis for Epigenetic Disruption in Cancer
The discovery of mutations in SMARCB1/SNF5 driving malignant
rhabdoid tumours first introduced genetic disruption of epigenetic control as a
mechanism of oncogenesis (Versteege et al., 1998),. Mutations in epigenetic
regulators continued to emerge from subsequent cancer studies, and have
surged in recent large-scale sequencing efforts (Figure 1-3). Epigenetic control
genes are mutated in about half of hepatocellular carcinomas (Fujimoto et al.,
2012) and bladder cancer (Gui et al., 2011) and represent six of the twelve most
significantly mutated genes in medulloblastoma (Pugh et al., 2012). It is
conceivable that disruption of epigenetic control by mutation of a key regulator
has the capacity to cause widespread changes to the transcriptome, multiplying
the effect of the single genetic alteration. It should be recognized that some of
the mutations reported for epigenetic regulators may be passenger events,
particularly in tumors with high background mutation rates. Therefore, we have
emphasized hotspot mutations and genes recurrently mutated at significant
frequencies. We focus here on somatic mutations, but germline variation have
also been shown to play a role in cancer. For example, germline mutations in
BAP1 have been found to be linked to a tumor predisposition syndrome
characterized by melanocytic tumors, mesothelioma, and uveal melanoma (Testa
et al., 2011; Wiesner et al., 2011), and rare germline allelic forms of PRDM9
have been found to be associated with childhood leukemia (Hussin et al., 2013).
20
Figure 1-3: Genetic Alterations in Epigenetic Regulators. Mutations and other genetic
alterations reported for selected epigenetic regulators are shown for various types of human
cancer. Malignancies are grouped by epithelial, hematological and other cancers. Mutations,
represented by colored cells, are deemed loss of function (red) unless evidence for gain of
function (either hypermorphic or neomorphic, blue) has been shown. Other genetic
alterations are plotted with different symbols, with a slash indicating translocation events and
a dot indicating copy number alterations. Translocations that generate oncogenic fusion
proteins, or neomorphic mutations that lead to oncogenic metabolites are represented in red
as well. Cells showing no entry may represent false negatives in our curation or in the
literature, and cancer types highly covered with whole-genome/exome studies (e.g. breast
cancer) might have fewer false negatives than those that are not. MSS/MSI – microsatellite
stable/instable; DLBCL - Diffuse large B-cell lymphoma; FL – follicular lymphoma.
21
DNA Methylation Writers and Erasers
The DNA methyltransferase DNMT3A is recurrently mutated in acute
myeloid leukemia (AML) and other myeloid malignancies (Ley et al., 2010; Yan et
al., 2011), as well as T-cell lymphoma (Couronne et al., 2012). The mutations
often occur at a R882 hotspot, but nevertheless likely reflect loss of function of
DNMT3A. Mutations in the DNA methylation eraser TET2 have also been
identified in the same cancer types (Abdel-Wahab et al., 2009; Langemeijer et al.,
2009; Quivoron et al., 2011), and bone marrow from patients with TET2
mutations show reduced levels of 5hmC (Ko et al., 2010). The isocitrate
dehydrogenases IDH1 and IDH2 are also recurrently mutated in AML. IDH1
enzymes with the R132 hotspot mutation and IDH2 enzymes containing R140 or
R172 mutations have lost the ability to produce alpha-ketoglutarate (α-KG), but
instead convert α-KG to an aberrant metabolite 2-hydroxyglutarate (2-HG), a
competitive inhibitor of alpha-ketoglutarate-dependent dioxygenases, such as the
TETs and JmjC-domain containing histone demethylases (Lu et al., 2012; Xu et
al., 2011). IDH1/2 mutations are mutually exclusive with TET2 mutations in AML,
consistent with the inhibitory effect of 2-HG on TETs as a mediator of the effects
of IDH1/2 mutations (Figueroa et al., 2010; Weissmann et al., 2012). The same
hotspot mutation for IDH1, and less often IDH2 is also found in gliomas and
glioblastomas. Both glioblastomas with IDH1 mutations (Noushmehr et al., 2010)
and cases of AML with mutation of IDH1 or IDH2 (Figueroa et al., 2010) display
CpG island methylator phenotypes.
Histone Gene Mutations
Mutations in histone variants H3.3 (H3F3A) or sometimes H3.1
(HIST1H3B) have been found in pediatric (Schwartzentruber et al., 2012; Wu et
al., 2012) and adult brain tumors (Sturm et al., 2012) with K27M and G34R or
22
G34V mutation hotspots. Tumors with G34 mutations display extensive DNA
hypomethylation, particularly in subtelomeric regions (Sturm et al., 2012),
perhaps contributing to alternative lengthening of telomeres (ALT)
(Schwartzentruber et al., 2012). Mutations were also observed in the ATRX and
DAXX genes, encoding proteins responsible for loading of the H3.3 variant into
the telomere region (Schwartzentruber et al., 2012). Pancreatic neuroendocrine
tumors (PanNETs) with ATRX and DAXX mutations also exhibit ALT (Heaphy et
al., 2011). Since this phenotypic effect is associated with a H3.3 loading defect,
the G34 mutations may also interfere with H3.3 loading. In contrast, tumors with
K27M mutations did not display ALT, and these mutations may instead mimic
dimethylated lysine 27, a repressive Polycomb mark, given that methionine is a
natural mimic of this epigenetic mark (Hyland et al., 2011). H3.3 G34R mutations
have also been reported in primitive neuroectodermal tumors of the CNS (Gessi
et al., 2013), mirroring the defect in the ATRX-DAXX-H3.3 axis in other brain
tumors and PanNETs.
Mutations in HIST1H3B and HIST1H1C have been found in diffuse large
B-cell lymphoma (DLBCL), although the mutations do not occur in clusters (Lohr
et al., 2012; Morin et al., 2011). These might be functionally different from the
hotspot mutations seen in brain tumors. Focal deletion of a histone gene cluster
at 6p22 is seen in near-haploid cases of acute lymphoblastic leukemia (Holmfeldt
et al., 2013).
Histone Methylation Writers
The MLL gene, encoding one of the H3K4 methyltransferases has over
50 translocation fusion partners in different lineages of leukemia. These
rearrangements account for 80% of the cases of infant leukemia and 5-10% of
adult leukemia cases, and are generally associated with poor prognosis (Tan et
23
al., 2011a). The primary mechanism has been attributed to the recruitment of
inappropriate epigenetic factors to MLL targets, by fusions between recruitment
proteins and the DNA-binding N-terminus of MLL. Target genes for these
recruited complexes include the HOX genes, particularly HOXA9, whose
upregulation is a key feature of MLL leukemia. MLL regulates the expression of
HOX genes in normal pluripotent cells, but the oncogenic fusion proteins keep
them from being turned off during differentiation and therefore impart stem-cell
like properties. Targeted therapeutic strategies are emerging for AML with MLL
fusions, including inhibition of menin (encoded by MEN1), DOT1L, PRMT1, the
histone acetylation reader BRD4, and LSD1 (Zeisig et al., 2012). In addition to
the translocations, loss-of-function mutations of MLL-MLL3 have been reported in
many different types of cancer, including AML - possibly another way of
disturbing the temporal control at promoters associated with pluripotency. MLL2
is mutated at very high frequency in B-cell follicular lymphoma and diffuse large
B-cell lymphoma, consistent with the gain-of-function mutations of EZH2 in the
same tumor types.
While menin is critical to the oncogenic effects of MLL fusion proteins in
AML (Yokoyama and Cleary, 2008), loss of function mutations have been found
in PanNETs (Jiao et al., 2011), consistent with a tumor-suppressor role,
suggesting that cellular context is important.
The recurrent t(5;11)(q35;p15.5) translocation in AML results in the fusion
of the H3K36 methyltransferase NSD1 to nucleoporin-98 (NUP98), with elevated
levels of H3K36me3 levels at HOXA genes and accompanying transcriptional
activation. Translocations involving another dedicated H3K36 methyltransferase
WHSC1/MMSET/NSD2 are seen in 20% of multiple myelomas. Another H3K36
24
methyltransferase, SETD2 is recurrently mutated in clear cell renal cell
carcinomas (ccRCC) (Dalgliesh et al., 2010).
A recent study reconstructed the phylogenetic structure of molecular
events in ccRCC with multiple spatially separated samples from the same tumors
(Gerlinger et al., 2012). In both of the two patients studied, distinct SETD2
inactivating mutations were found in different parts of the same tumor.
Immunohistochemistry staining confirmed H3K36me3 loss in all the mutant
tumors. This convergent somatic evolution indicates that failure to establish
H3K36 methylation marks provides a strong selective advantage relatively late in
ccRCC progression. A similar molecular convergence was found for KDM5C, an
H3K4 demethylase, in one of the two patients. This, together with recurrent
mutations in other epigenetic regulators, shows that epigenetic dysregulation,
often mediated by genetic events, is important in advanced ccRCCs.
EZH2, the writer for the H3K27 methylation mark associated with
Polycomb repression, has long been viewed as an oncogene in cancer. Indeed,
gain-of-function mutations are seen in lymphomas. However, loss-of-function
mutations in this gene have recently been described in other cancers. We
discuss these divergent effects of EZH2 mutations and other alterations to this
pathway in more detail later (Figure 1-4).
Histone Methylation Erasers
Consistent with EZH2 overexpression in various solid tumors, the
corresponding eraser KDM6A/UTX is mutated in more than a dozen tumor types,
with the highest frequency in bladder (Gui et al., 2011; van Haaften et al., 2009).
The H3K9 demethylase KDM4C/GASC1 is amplified in breast cancer and has
been shown to drive transformation (Liu et al., 2009; Rui et al., 2010). Ectopic
expression of this putative oncogene in vitro causes an efficient decrease of
25
H3K9me3 (Cloos et al., 2006). Its co-amplification with JAK2 (Rui et al., 2010) -
which phosphorylates H3Y41 and prevents binding of H3K9 methylation reader
HP1 to the H3K9 methylation mark - in lymphoma makes for an interesting
example of a single genetic event hitting two possible epigenetic regulators.
Inhibiting the two co-amplified and cooperating gene products is efficient at killing
these lymphoma cells.
Histone Acetylation Writers and Erasers
The counteracting histone acetyltransferases (HATs) and deacetylases
(HDACs) are considered to be promiscuous, and often have important non-
histone substrates such as TP53. There are three major families of HATs,
namely the CBP/P300, GNAT, and the MYST families. CREBBP is mutated at
high frequency in follicular lymphoma and diffuse large-cell B-cell lymphoma
(DLBCL) (Morin et al., 2011; Pasqualucci et al., 2011) and in ALL (Mullighan et
al., 2011), particularly relapsed hyperdiploid ALL (Inthal et al., 2012). Its
paralogue EP300 also undergoes frequent mutation (Gayther et al., 2000; Gui et
al., 2011; Pasqualucci et al., 2011; Zhang et al., 2012) and loss of heterozygosity
(LOH) in many different epithelial cancers.
HATs have also been implicated in gene fusions. The t(8;16)(p11;p13)
translocation in AML fuses the N-terminal part of MOZ, the founding member of
the MYST HAT family, to the major part of the CBP gene containing the
acetylase domain. MOZ has also been found to be involved in fusions with
EP300 (Yang, 2004). These translocations generating chimeric oncoproteins with
the DNA-binding domain of MOZ and the transcription-activating domain of
another coactivator are associated with AML M5/M4. These results suggest that
both disruption and redirection of HAT could contribute to cancer.
26
Reports of mutations in HDACs are rare. Rather, HDACs are often co-
opted by other genetic alterations. A prime example is the PML-RARα
translocation, responsible for 95% of the AML FAB-M3 (APL, Acute
promyelocytic leukemia) cases. The leukemogenetic effect of this translocation is
primarily mediated through aberrant recruitment of N-CoR/HDAC repressor
complexes (Minucci and Pelicci, 2006). The retinoic acid receptor-alpha (RARα)
part binds to retinoic acid-responsive elements (RAREs), while the PML moiety
recruits the HDAC-containing repressive complex. All-trans retinoic acid (ATRA)
targets RARα, dissociates these repressor complexes and effectively induces
differentiation of the leukemic promyelocytes. Combination therapy of ATRA and
arsenic trioxide shows excellent clinical response, and turned APL into a highly
curable disease (Wang and Chen, 2008). Another fusion protein PLZF-RARα
blocks differentiation by a similar mechanism. However, APL with this
translocation is ATRA-resistant due to the higher affinity of the PLZF moiety to
the N-CoR complex, but a combination of ATRA with HDAC inhibitors can fully
reverse the transcriptional repression and induce terminal differentiation for this
type of AML (Wang and Chen, 2008). Similarly, the RUNX1-ETO fusion, the
AML1-ETO fusion, and the CBF-MYH11 protein from inv(16), all recruit HDACs,
and efficacy of HDAC inhibitors has been demonstrated for all three (Zeisig et al.,
2012).
Epigenetic Readers
The epigenetic readers add another layer of control to the epigenetic
state, by serving as interpreters of the epigenetic state and relaying epigenetic
signals. Many of the epigenetic writer/eraser/remodelers have intrinsic reader
domains, or interact with dedicated readers to sense the presence or absence of
particular epigenetic marks. Translocations joining BRD4 or occasionally BRD3,
27
both readers of the BET Bromodomain-containing family, to almost the entire
length of the NUT gene define a lethal, poorly differentiated pediatric tumor, NUT
midline carcinoma (NMC) (French et al., 2008). The BET family members
targetable by BET inhibitors (Filippakopoulos et al., 2010), are lysine acetylation
readers that bind transcriptionally active chromatin as acetylated lysine readers,
and are targetable by BET inhibitors (Filippakopoulos et al., 2010). Functional
studies show that the BRD-NUT fusion oncoprotein binds avidly to acetylated
histones, resulting in a differentiation block, potentially by interfering with
transcriptional programs driving differentiation (French et al., 2008). BRD3 is
significantly mutated in lung adenocarcinomas (Imielinski et al., 2012), and BRD8
in liver cancer (Fujimoto et al., 2012). In addition, the plant homeodomain (PHD)-
domain containing gene PHF6 is recurrently mutated in AML (Van Vlierberghe et
al., 2011) and overall loss of this gene (mutation and/or deletion) is observed in
T-ALL (Van Vlierberghe et al., 2010).
Chromatin Remodelers
A large number of SWI/SNF complexes exist in mammals and contribute
to lineage- and tissue-specific gene expression (Wilson and Roberts, 2011). The
two major types of SWI/SNF complexes are BAF (BRM-containing, or SWI/SNF-
A) and PBAF (BRG1 containing, or SWI/SNF-B), defined by a core enzymatic
unit being either SMARCA2 (BRM) or SMARCA4 (BRG1). Those complexes also
contain other core units, such as SMARCB1/SNF5 and ARID1A/B, unique to
BAF, and PBRM1 and BRD7, unique to PBAF. Truncating mutations in the
SMARCB1 gene are very common in malignant rhabdoid tumours (RTs)
(Versteege et al., 1998), a rare yet lethal tumor diagnosed in children. The
biallelic nature of the inactivation fits with a tumor suppressor role for SMARCB1.
Familial cases of RTs are associated with inheritance of one defective
28
SMARCB1 allele. SMARCB1 is also mutated in a few other cancers (Figure 1-1).
SMARCA4 mutation is also seen in familial cases of RT (Schneppenheim et al.,
2010), indicating that it is indeed SWI/SNF dysfunction that is responsible for RT
development. SMARCA4 is also mutated Burkitt’s Lymphoma, in a mutually
exclusive manner with ARID1A mutations (Love et al., 2012), again suggesting a
driver role for SWI/SNF mutations. Recurrent SMARCA4 mutation is also seen in
lung cancer (Imielinski et al., 2012) and medulloblastoma (especially the WNT
subtype) (Jones et al., 2012; Parsons et al., 2011; Pugh et al., 2012; Robinson et
al., 2012).
ARID1A mutations have been found in more than ten different tumor
types, with the highest rate in the clear cell subtype of ovarian cancer, where it is
mutated in more than half of the tumors (Jones et al., 2010; Wiegand et al.,
2010). ARID1A is also mutated in the endometrioid subtypes of ovarian
(Wiegand et al., 2010) and endometrial cancers (Guan et al., 2011). ARID1B and
ARID2 mutations are also seen in various cancer types, including liver cancer
(Fujimoto et al., 2012) and melanoma (Hodis et al., 2012), among others. In
addition, the polybromo-containing PBRM1 in the PBAF complex was recently
found to be the second most mutated gene in clear cell renal cell carcinomas
(Varela et al., 2011). Another SWI/SNF gene, SMARCE1, is highly recurrently
mutated in clear cell meningiomas (Smith et al., 2013). The exceptionally high
mutation rate of SWI/SNF member in clear cell tumors from different tissues
(ovary, kidney, and meninges) highlights an interesting possible link between
clear cell tumors and SWI/SNF dysfunction.
ATRX, responsible for H3.3 incorporation at telomeres and pericentric
heterochromatin is often mutated in pancreatic neuroendocrine tumours
(PanNETs), the second most common malignancy of the pancreas (Jiao et al.,
29
2011). Interestingly, there are also recurrent mutations in the associated
chaperone DAXX in the same cancer type, and the two mutations are mutually
exclusive. Mutation of these two genes are also found in GBM where they are
mutually exclusive with the H3F3A mutations described earlier, with any of these
three genes mutated in almost half of the tumors studied (Schwartzentruber et al.,
2012). These mutations all lead to alternative lengthening of telomeres
associated with increased genomic instability (Heaphy et al., 2011;
Schwartzentruber et al., 2012). With the possible exception of ATRX and DAXX,
most of the mutations in the SWI/SNF family members are not associated with
genomic instability (Wilson and Roberts, 2011). Rather, perturbed differentiation
may be the major mechanism, as cells from different lineages co-exist within an
individual rhabdoid tumor.
The CHD family chromatin remodelers can be divided into three classes.
Class I (CHD1/2), Class II (CHD3/4, in the NuRD/Mi-2/CHD complex), and Class
III (CHD5-9). The NuRD/Mi-2/CHD complexes are unique in that they have core
enzymatic subunits with at least two distinct functions: ATP-dependent
remodeling (CHD3 and CHD4), as well as histone deacetylase (HDAC1 and
HDAC2) functions (Lai and Wade, 2011), and therefore couple two epigenetic
processes in one complex for transcriptional repression. Their MBD and MTA
subunits target the complex to different parts of the genome by binding
methylated DNA (MBD) or other transcription factors (MTA), in physiological and
pathological conditions. For example, the MTA-2 containing NuRD complex
associates with TWIST in breast cancer, represses genes such as E-cadherin
(CDH1), and contributes to EMT. CHD4 is mutated in 17% of serous endometrial
cancer (Le Gallo et al., 2012). Of the other CHDs, CHD1 is the second most-
frequently deleted gene in prostate cancer, defining an ETS-negative subtype
30
(Grasso et al., 2012), with mutations reported as well (Berger et al., 2011).
Epigenetic Insulators
CTCF is located in 16q22.1 with LOH in breast and prostate cancers
(Filippova et al., 1998) and Wilms’ tumors (Mummert et al., 2005). CTCF
mutation has also been reported in breast cancer (Filippova et al., 2002), and this
mutation is found to be significant in a cohort of 510 tumors (TCGA, 2012c). Rare
mutations in this gene have also been reported for prostate cancer (Filippova et
al., 2002), Wilms’ tumor (Filippova et al., 2002), AML (Dolnik et al., 2012), ALL
(Mullighan et al., 2011; Zhang et al., 2012)), and endometrial cancers (Le Gallo
et al., 2012). The functional implications of CTCF deletion/mutation, especially
given the low frequency, have not been fully delineated yet, but abrogation of
proper insulation might be one mechanism.
Genetic Disruption of a Central Epigenetic Control Circuit
Figure 1-4 illustrates the diverse ways in which a central epigenetic
control circuit can be impacted in cancer. EZH2 catalyzes methylation at H3K27,
as part of PRC2. Two other core subunits of PRC2 are EED and SUZ12, and
other components such as JARID2 can be part of a PRC2 complex too. EZH2
has long been thought to be oncogenic since it is overexpressed as a result of
amplification of EZH2 in breast, bladder and other cancers (Bracken et al., 2003),
as well as genetic loss of miR101, which represses EZH2 in prostate cancer
(Varambally et al., 2008). In line with this view, gain-of-function hotspot mutations
(Y641 and A677) in the SET-domain of EZH2 have been found in a significant
portion of lymphomas (Lohr et al., 2012; Morin et al., 2010; Morin et al., 2011;
Pasqualucci et al., 2011). More convincingly, EZH2 amplification and
overexpression in two of the largest subgroups of medulloblastoma is mutually
31
exclusive with mutation of the H3K27 demethylase KDM6A, suggesting that
accumulation of H3K27me3 is a key step in these tumors (Robinson et al., 2012).
On the other hand, loss of function mutations of EZH2 have also been
found in a series of myeloid malignancies, including MDS, Multiple Myeloma,
MPN and MDS/MPN, as well as in head-and-neck squamous cell carcinomas
(HNSCCs) (Ernst et al., 2010; Nikoloski et al., 2010; Stransky et al., 2011),
suggesting that PRC2 can also act as tumor suppressors. Aside from the
myeloid malignancies, mutually exclusive recurrent deletion and loss of function
mutations of EZH2 and SUZ12 have also been found in T-lineage acute
lymphoblastic leukemia (T-ALL) (Ntziachristos et al., 2012) or of all three PRC2
subunits in early-T-cell-precursor ALL (Zhang et al., 2012) further substantiate a
tumor suppressor role. Indeed, disruption of EZH2 is sufficient to induce T-ALL in
mice (Simon et al., 2012). PRC2 component mutations are much more common
in early T-cell precursor ALL, a lymphoblastic leukemia with myeloid features.
The Polycomb repressive deubiquitinase (PR-DUB) component ASXL1 is also
mutated in myeloid malignancies, and ASXL1 mutation mediates myeloid
transformation through loss of PRC2 repression (Abdel-Wahab et al., 2012), in
line with the observed loss of function of PRC2 members in myeloid disorders.
Mouse models also show that loss of BAP1, the enzymatic unit of PR-DUB, lead
to myeloid transformation (Dey et al., 2012). This, together with a BAP1 catalytic
mutation found in a MDS patient lacking other MDS mutations (Dey et al., 2012)
further lends credibility to the idea that loss of Polycomb repression drive myeloid
disorders, while in B-cell lymphoma and solid tumors gain of Polycomb
repression seem important.
EZH2 also exhibits PRC2-independent oncogenic activities. For example,
in castration-resistant prostate cancer Akt-mediated phosphorylation of EZH2 at
32
S21 can shift EZH2 from PRC2-dependent promoters to EZH2 ‘solo’ promoters.
This EZH2 activity is often associated with androgen receptor (AR), and activates
gene expression at these loci (Xu et al., 2012).
These complex ways in which the H3K27me3 axis is disrupted in cancer
suggest differential therapeutic approaches for different cancer types. In
particular, caution should be used when considering EZH2 inhibitors for the
myeloid malignancies featuring genetic lesions leading to PRC2 repression loss.
Figure 1-4: Genetic Disruption of Epigenetic Control at H3K27 in Cancer. The
counteracting writer and eraser EZH2 and KDM6A/UTX form a pair in regulating an important
epigenetic mark, methylation at H3 lysine 27. EZH2 catalyzes the methylation process with
help from other components in the Polycomb Repressive Complex 2 (PRC2), while KDM6A,
part of the Trithorax complex, removes this repressive mark. The K27me3 mark attracts
another Polycomb complex, PRC1, which ubiquitinates H2AK119, and thereby blocks PolII
elongation. Another Polycomb complex, PR-DUB is also critical to the maintainance of the
repression at a subset of the Polycomb genes, although it removes the H2AK119ub mark
and thus counteracts PRC1 in that regard. Mutations and genetic alterations spanning a wide
spectrum of human cancers hit this epigenetic pathway. Solid tumors show possible
neomorphic histone K27 methylation, UTX mutation, EZH2 amplification and/or
overexpression due to genomic loss of the repressive microRNA miR101, as well as
amplification/overexpression of the PRC1 member BMI1, and lymphoma exhibits gain-of-
function mutations of EZH2, consistent with a gain of Polycomb repression (red boxes) in the
affected malignancies. In contrast, myeloid malignancies and ALL, particularly early T-cell
precursor ALL show mutations that could sabotage Polycomb repression (blue boxes). Gray
boxes indicate that the effect on K27me3 is not clear.
33
Conclusions
Given the importance of epigenetics in cancer, we set out to study
epigenetic disruption in three types of human cancer, aiming at defining
epigenetic subtypes such as CIMP, identifying epigenetically silenced genes in a
high-throughput manner, and delineating the influence of genetics on epigenetics
or vice versa. We have done this in the context of the Cancer Genome Atlas. We
have also taken one candidate gene that came out of such analyses, and tested
it as a candidate for germline susceptibility gene for serous ovarian cancer.
34
Chapter One Bibliogaphy
Abdel-Wahab, O., Adli, M., LaFave, L.M., Gao, J., Hricik, T., Shih, A.H., Pandey,
S., Patel, J.P., Chung, Y.R., Koche, R., et al. (2012). ASXL1 mutations promote
myeloid transformation through loss of PRC2-mediated gene repression. Cancer
Cell 22, 180-193.
Abdel-Wahab, O., Mullally, A., Hedvat, C., Garcia-Manero, G., Patel, J.,
Wadleigh, M., Malinge, S., Yao, J., Kilpivaara, O., Bhat, R., et al. (2009). Genetic
characterization of TET1, TET2, and TET3 alterations in myeloid malignancies.
Blood 114, 144-147.
Adams, D., Altucci, L., Antonarakis, S.E., Ballesteros, J., Beck, S., Bird, A., Bock,
C., Boehm, B., Campo, E., Caricasole, A., et al. (2012). BLUEPRINT to decode
the epigenetic signature written in blood. Nat Biotechnol 30, 224-226.
Baer, C., Claus, R., and Plass, C. (2013). Genome-Wide Epigenetic Regulation
of miRNAs in Cancer. Cancer Res 73, 473-477.
Baylin, S.B., Hoppener, J.W., de Bustros, A., Steenbergh, P.H., Lips, C.J., and
Nelkin, B.D. (1986). DNA methylation patterns of the calcitonin gene in human
lung cancers and lymphomas. Cancer Res 46, 2917-2922.
Baylin, S.B., and Jones, P.A. (2011). A decade of exploring the cancer
epigenome - biological and translational implications. Nat Rev Cancer 11, 726-
734.
Berger, M.F., Lawrence, M.S., Demichelis, F., Drier, Y., Cibulskis, K., Sivachenko,
A.Y., Sboner, A., Esgueva, R., Pflueger, D., Sougnez, C., et al. (2011). The
genomic complexity of primary human prostate cancer. Nature 470, 214-220.
Berman, B.P., Weisenberger, D.J., Aman, J.F., Hinoue, T., Ramjan, Z., Liu, Y.,
Noushmehr, H., Lange, C.P., van Dijk, C.M., Tollenaar, R.A., et al. (2012).
Regions of focal DNA hypermethylation and long-range hypomethylation in
colorectal cancer coincide with nuclear lamina-associated domains. Nature
Genetics 44, 40-46.
Bernstein, B.E., Mikkelsen, T.S., Xie, X., Kamal, M., Huebert, D.J., Cuff, J., Fry,
B., Meissner, A., Wernig, M., Plath, K., et al. (2006). A bivalent chromatin
structure marks key developmental genes in embryonic stem cells. Cell 125, 315-
326.
35
Bird, A., Taggart, M., Frommer, M., Miller, O.J., and Macleod, D. (1985). A
fraction of the mouse genome that is derived from islands of nonmethylated,
CpG-rich DNA. Cell 40, 91-99.
Boyarchuk, E., Montes de Oca, R., and Almouzni, G. (2011). Cell cycle dynamics
of histone variants at the centromere, a model for chromosomal landmarks. Curr
Opin Cell Biol 23, 266-276.
Bracken, A.P., Pasini, D., Capra, M., Prosperini, E., Colli, E., and Helin, K. (2003).
EZH2 is downstream of the pRB-E2F pathway, essential for proliferation and
amplified in cancer. EMBO J 22, 5323-5335.
Cedar, H., and Bergman, Y. (2012). Programming of DNA methylation patterns.
Annu Rev Biochem 81, 97-117.
Challen, G.A., Sun, D., Jeong, M., Luo, M., Jelinek, J., Berg, J.S., Bock, C.,
Vasanthakumar, A., Gu, H., Xi, Y., et al. (2012). Dnmt3a is essential for
hematopoietic stem cell differentiation. Nat Genet 44, 23-31.
Chi, P., Allis, C.D., and Wang, G.G. (2010). Covalent histone modifications--
miswritten, misinterpreted and mis- erased in human cancers. Nat Rev Cancer
10, 457-469.
Clark, S.J. (2007). Action at a distance: epigenetic silencing of large
chromosomal regions in carcinogenesis. Hum Mol Genet 16, R88-R95.
Cloos, P.A., Christensen, J., Agger, K., Maiolica, A., Rappsilber, J., Antal, T.,
Hansen, K.H., and Helin, K. (2006). The putative oncogene GASC1
demethylates tri- and dimethylated lysine 9 on histone H3. Nature 442, 307-311.
Cohen, N.M., Kenigsberg, E., and Tanay, A. (2011). Primate CpG islands are
maintained by heterogeneous evolutionary regimes involving minimal selection.
Cell 145, 773-786.
Coolen, M.W., Stirzaker, C., Song, J.Z., Statham, A.L., Kassir, Z., Moreno, C.S.,
Young, A.N., Varma, V., Speed, T.P., Cowley, M., et al. (2010). Consolidation of
the cancer genome into domains of repressive chromatin by long-range
epigenetic silencing (LRES) reduces transcriptional plasticity. Nat Cell Biol 12,
235-246.
Couronne, L., Bastard, C., and Bernard, O.A. (2012). TET2 and DNMT3A
mutations in human T-cell lymphoma. N Engl J Med 366, 95-96.
36
Craene, B.D., and Berx, G. (2012). Regulatory networks defining EMT during
cancer initiation and progression. Nat Rev Cancer 13, 97-110.
Dalgliesh, G.L., Furge, K., Greenman, C., Chen, L., Bignell, G., Butler, A., Davies,
H., Edkins, S., Hardy, C., Latimer, C., et al. (2010). Systematic sequencing of
renal carcinoma reveals inactivation of histone modifying genes. Nature 463,
360-363.
Dawson, M.A., and Kouzarides, T. (2012). Cancer epigenetics: from mechanism
to therapy. Cell 150, 12-27.
Daxinger, L., and Whitelaw, E. (2012). Understanding transgenerational
epigenetic inheritance via the gametes in mammals. Nat Rev Genet 13, 153-162.
De Carvalho, D.D., Sharma, S., You, J.S., Su, S.F., Taberlay, P.C., Kelly, T.K.,
Yang, X., Liang, G., and Jones, P.A. (2012). DNA methylation screening
identifies driver epigenetic events of cancer cell survival. Cancer Cell 21, 655-
667.
De, S., and Michor, F. (2011). DNA secondary structures and epigenetic
determinants of cancer genome evolution. Nat Struct Mol Biol 18, 950-955.
Dey, A., Seshasayee, D., Noubade, R., French, D.M., Liu, J., Chaurushiya, M.S.,
Kirkpatrick, D.S., Pham, V.C., Lill, J.R., Bakalarski, C.E., et al. (2012). Loss of the
tumor suppressor BAP1 causes myeloid transformation. Science 337, 1541-1546.
Diala, E.S., and Hoffman, R.M. (1982). Hypomethylation of HeLa cell DNA and
the absence of 5-methylcytosine in SV40 and adenovirus (type 2) DNA: analysis
by HPLC. Biochem Biophys Res Commun 107, 19-26.
Dion, V., Lin, Y., Hubert, L., Jr., Waterland, R.A., and Wilson, J.H. (2008). Dnmt1
Deficiency Promotes CAG Repeat Expansion in the Mouse Germline. Hum Mol
Genet 5, 5.
Dolnik, A., Engelmann, J.C., Scharfenberger-Schmeer, M., Mauch, J.,
Kelkenberg-Schade, S., Haldemann, B., Fries, T., Kronke, J., Kuhn, M.W.,
Paschka, P., et al. (2012). Commonly altered genomic regions in acute myeloid
leukemia are enriched for somatic mutations involved in chromatin remodeling
and splicing. Blood 120, e83-92.
Easwaran, H., Johnstone, S.E., Van Neste, L., Ohm, J., Mosbruger, T., Wang, Q.,
Aryee, M.J., Joyce, P., Ahuja, N., Weisenberger, D., et al. (2012). A DNA
37
hypermethylation module for the stem/progenitor cell signature of cancer.
Genome Res 22, 837-849.
Eden, A., Gaudet, F., Waghmare, A., and Jaenisch, R. (2003). Chromosomal
instability and tumors promoted by DNA hypomethylation. Science 300, 455.
Ehrlich, M., Gama-Sosa, M.A., Huang, L.H., Midgett, R.M., Kuo, K.C., McCune,
R.A., and Gehrke, C. (1982). Amount and distribution of 5-methylcytosine in
human DNA from different types of tissues of cells. Nucleic Acids Res 10, 2709-
2721.
Ehrlich, M., and Wang, R.Y. (1981). 5-Methylcytosine in eukaryotic DNA. Science
212, 1350-1357.
Encode-Project-Consortium, Dunham, I., Kundaje, A., Aldred, S.F., Collins, P.J.,
Davis, C.A., Doyle, F., Epstein, C.B., Frietze, S., Harrow, J., et al. (2012). An
integrated encyclopedia of DNA elements in the human genome. Nature 489, 57-
74.
Ernst, J., Kheradpour, P., Mikkelsen, T.S., Shoresh, N., Ward, L.D., Epstein, C.B.,
Zhang, X., Wang, L., Issner, R., Coyne, M., et al. (2011). Mapping and analysis
of chromatin state dynamics in nine human cell types. Nature 473, 43-49.
Ernst, T., Chase, A.J., Score, J., Hidalgo-Curtis, C.E., Bryant, C., Jones, A.V.,
Waghorn, K., Zoi, K., Ross, F.M., Reiter, A., et al. (2010). Inactivating mutations
of the histone methyltransferase gene EZH2 in myeloid disorders. Nat Genet 42,
722-726.
Estecio, M.R., Gallegos, J., Vallot, C., Castoro, R.J., Chung, W., Maegawa, S.,
Oki, Y., Kondo, Y., Jelinek, J., Shen, L., et al. (2010). Genome architecture
marked by retrotransposons modulates predisposition to DNA methylation in
cancer. Genome Res 20, 1369-1382.
Esteller, M., Garcia-Foncillas, J., Andion, E., Goodman, S.N., Hidalgo, O.F.,
Vanaclocha, V., Baylin, S.B., and Herman, J.G. (2000a). Inactivation of the DNA-
repair gene MGMT and the clinical response of gliomas to alkylating agents. N
Engl J Med 343, 1350-1354.
Esteller, M., Risques, R.A., Toyota, M., Capella, G., Moreno, V., Peinado, M.A.,
Baylin, S.B., and Herman, J.G. (2001). Promoter hypermethylation of the DNA
repair gene O(6)-methylguanine-DNA methyltransferase is associated with the
presence of G:C to A:T transition mutations in p53 in human colorectal
tumorigenesis. Cancer Res 61, 4689-4692.
38
Esteller, M., Toyota, M., Sanchez-Cespedes, M., Capella, G., Peinado, M.A.,
Watkins, D.N., Issa, J.P., Sidransky, D., Baylin, S.B., and Herman, J.G. (2000b).
Inactivation of the DNA repair gene O6-methylguanine-DNA methyltransferase
by promoter hypermethylation is associated with G to A mutations in K-ras in
colorectal tumorigenesis. Cancer Res 60, 2368-2371.
Fabbri, M., and Calin, G.A. (2010). Epigenetics and miRNAs in human cancer.
Adv Genet 70, 87-99.
Feinberg, A.P., and Vogelstein, B. (1983). Hypomethylation distinguishes genes
of some human cancers from their normal counterparts. Nature 301, 89-92.
Fenouil, R., Cauchy, P., Koch, F., Descostes, N., Cabeza, J.Z., Innocenti, C.,
Ferrier, P., Spicuglia, S., Gut, M., Gut, I., et al. (2012). CpG islands and GC
content dictate nucleosome depletion in a transcription-independent manner at
mammalian promoters. Genome Res 22, 2399-2408.
Figueroa, M.E., Abdel-Wahab, O., Lu, C., Ward, P.S., Patel, J., Shih, A., Li, Y.,
Bhagwat, N., Vasanthakumar, A., Fernandez, H.F., et al. (2010). Leukemic IDH1
and IDH2 mutations result in a hypermethylation phenotype, disrupt TET2
function, and impair hematopoietic differentiation. Cancer Cell 18, 553-567.
Filippakopoulos, P., Qi, J., Picaud, S., Shen, Y., Smith, W.B., Fedorov, O., Morse,
E.M., Keates, T., Hickman, T.T., Felletar, I., et al. (2010). Selective inhibition of
BET bromodomains. Nature 468, 1067-1073.
Filippova, G.N., Lindblom, A., Meincke, L.J., Klenova, E.M., Neiman, P.E., Collins,
S.J., Doggett, N.A., and Lobanenkov, V.V. (1998). A widely expressed
transcription factor with multiple DNA sequence specificity, CTCF, is localized at
chromosome segment 16q22.1 within one of the smallest regions of overlap for
common deletions in breast and prostate cancers. Genes Chromosomes Cancer
22, 26-36.
Filippova, G.N., Qi, C.F., Ulmer, J.E., Moore, J.M., Ward, M.D., Hu, Y.J.,
Loukinov, D.I., Pugacheva, E.M., Klenova, E.M., Grundy, P.E., et al. (2002).
Tumor-associated zinc finger mutations in the CTCF transcription factor
selectively alter tts DNA-binding specificity. Cancer Res 62, 48-52.
Fraga, M.F., Ballestar, E., Villar-Garea, A., Boix-Chornet, M., Espada, J., Schotta,
G., Bonaldi, T., Haydon, C., Ropero, S., Petrie, K., et al. (2005). Loss of
acetylation at Lys16 and trimethylation at Lys20 of histone H4 is a common
hallmark of human cancer. Nat Genet 37(4), 391-400.
39
French, C.A., Ramirez, C.L., Kolmakova, J., Hickman, T.T., Cameron, M.J.,
Thyne, M.E., Kutok, J.L., Toretsky, J.A., Tadavarthy, A.K., Kees, U.R., et al.
(2008). BRD-NUT oncoproteins: a family of closely related nuclear proteins that
block epithelial differentiation and maintain the growth of carcinoma cells.
Oncogene 27, 2237-2242.
Fujimoto, A., Totoki, Y., Abe, T., Boroevich, K.A., Hosoda, F., Nguyen, H.H., Aoki,
M., Hosono, N., Kubo, M., Miya, F., et al. (2012). Whole-genome sequencing of
liver cancers identifies etiological influences on mutation patterns and recurrent
mutations in chromatin regulators. Nat Genet 44, 760-764.
Gaffney, D.J., McVicker, G., Pai, A.A., Fondufe-Mittendorf, Y.N., Lewellen, N.,
Michelini, K., Widom, J., Gilad, Y., and Pritchard, J.K. (2012). Controls of
nucleosome positioning in the human genome. PLoS Genet 8, e1003036.
Gal-Yam, E.N., Egger, G., Iniguez, L., Holster, H., Einarsson, S., Zhang, X., Lin,
J.C., Liang, G., Jones, P.A., and Tanay, A. (2008). Frequent switching of
Polycomb repressive marks and DNA hypermethylation in the PC3 prostate
cancer cell line. Proc Natl Acad Sci U S A 105, 12979-12984.
Gayther, S.A., Batley, S.J., Linger, L., Bannister, A., Thorpe, K., Chin, S.F.,
Daigo, Y., Russell, P., Wilson, A., Sowter, H.M., et al. (2000). Mutations
truncating the EP300 acetylase in human cancers. Nat Genet 24, 300-303.
Gerlinger, M., Rowan, A.J., Horswell, S., Larkin, J., Endesfelder, D., Gronroos, E.,
Martinez, P., Matthews, N., Stewart, A., Tarpey, P., et al. (2012). Intratumor
heterogeneity and branched evolution revealed by multiregion sequencing. N
Engl J Med 366, 883-892.
Gertz, J., Varley, K.E., Reddy, T.E., Bowling, K.M., Pauli, F., Parker, S.L., Kucera,
K.S., Willard, H.F., and Myers, R.M. (2011). Analysis of DNA methylation in a
three-generation family reveals widespread genetic influence on epigenetic
regulation. PLoS genetics 7, e1002228.
Gessi, M., Gielen, G.H., Hammes, J., Dorner, E., Muhlen, A.Z., Waha, A., and
Pietsch, T. (2013). H3.3 G34R mutations in pediatric primitive neuroectodermal
tumors of central nervous system (CNS-PNET) and pediatric glioblastomas:
possible diagnostic and therapeutic implications? J Neurooncol.
Ginno, P.A., Lott, P.L., Christensen, H.C., Korf, I., and Chedin, F. (2012). R-loop
formation is a distinctive characteristic of unmethylated human CpG island
promoters. Mol Cell 45, 814-825.
40
Gonzalo, S., Jaco, I., Fraga, M.F., Chen, T., Li, E., Esteller, M., and Blasco, M.A.
(2006). DNA methyltransferases control telomere length and telomere
recombination in mammalian cells. Nat Cell Biol 8, 416-424.
Graff, J.R., Gabrielson, E., Fujii, H., Baylin, S.B., and Herman, J.G. (2000).
Methylation patterns of the E-cadherin 5' CpG island are unstable and reflect the
dynamic, heterogeneous loss of E-cadherin expression during metastatic
progression. J Biol Chem 275, 2727-2732.
Grasso, C.S., Wu, Y.M., Robinson, D.R., Cao, X., Dhanasekaran, S.M., Khan,
A.P., Quist, M.J., Jing, X., Lonigro, R.J., Brenner, J.C., et al. (2012). The
mutational landscape of lethal castration-resistant prostate cancer. Nature 487,
239-243.
Guan, B., Mao, T.L., Panuganti, P.K., Kuhn, E., Kurman, R.J., Maeda, D., Chen,
E., Jeng, Y.M., Wang, T.L., and Shih Ie, M. (2011). Mutation and loss of
expression of ARID1A in uterine low-grade endometrioid carcinoma. Am J Surg
Pathol 35, 625-632.
Gui, Y., Guo, G., Huang, Y., Hu, X., Tang, A., Gao, S., Wu, R., Chen, C., Li, X.,
Zhou, L., et al. (2011). Frequent mutations of chromatin remodeling genes in
transitional cell carcinoma of the bladder. Nat Genet 43, 875-878.
Guo, G., Wang, W., and Bradley, A. (2004). Mismatch repair genes identified
using genetic screens in Blm-deficient embryonic stem cells. Nature 429, 891-
895.
Hanahan, D., and Weinberg, R.A. (2011). Hallmarks of cancer: the next
generation. Cell 144, 646-674.
Hansen, K.D., Timp, W., Bravo, H.C., Sabunciyan, S., Langmead, B., McDonald,
O.G., Wen, B., Wu, H., Liu, Y., Diep, D., et al. (2011). Increased methylation
variation in epigenetic domains across cancer types. Nature genetics 43, 768-
775.
Hawkins, R.D., Hon, G.C., Lee, L.K., Ngo, Q., Lister, R., Pelizzola, M., Edsall,
L.E., Kuan, S., Luu, Y., Klugman, S., et al. (2010). Distinct epigenomic
landscapes of pluripotent and lineage-committed human cells. Cell Stem Cell 6,
479-491.
Heaphy, C.M., de Wilde, R.F., Jiao, Y., Klein, A.P., Edil, B.H., Shi, C.,
Bettegowda, C., Rodriguez, F.J., Eberhart, C.G., Hebbar, S., et al. (2011).
Altered telomeres in tumors with ATRX and DAXX mutations. Science 333, 425.
41
Hegi, M.E., Diserens, A.C., Gorlia, T., Hamou, M.F., de Tribolet, N., Weller, M.,
Kros, J.M., Hainfellner, J.A., Mason, W., Mariani, L., et al. (2005). MGMT gene
silencing and benefit from temozolomide in glioblastoma. N Engl J Med 352(10),
997-1003.
Herman, J.G., Umar, A., Polyak, K., Graff, J.R., Ahuja, N., Issa, J.P., Markowitz,
S., Willson, J.K., Hamilton, S.R., Kinzler, K.W., et al. (1998). Incidence and
functional consequences of hMLH1 promoter hypermethylation in colorectal
carcinoma. Proc Natl Acad Sci U S A 95, 6870-6875.
Hinoue, T., Weisenberger, D.J., Lange, C.P., Shen, H., Byun, H.M., Van Den
Berg, D., Malik, S., Pan, F., Noushmehr, H., van Dijk, C.M., et al. (2012).
Genome-scale analysis of aberrant DNA methylation in colorectal cancer.
Genome research 22, 271-282.
Hinoue, T., Weisenberger, D.J., Pan, F., Campan, M., Kim, M., Young, J.,
Whitehall, V.L., Leggett, B.A., and Laird, P.W. (2009). Analysis of the association
between CIMP and BRAF in colorectal cancer by DNA methylation profiling.
PLoS One 4, e8357.
Hitchins, M.P., Rapkins, R.W., Kwok, C.T., Srivastava, S., Wong, J.J.,
Khachigian, L.M., Polly, P., Goldblatt, J., and Ward, R.L. (2011). Dominantly
inherited constitutional epigenetic silencing of MLH1 in a cancer-affected family is
linked to a single nucleotide variant within the 5'UTR. Cancer Cell 20, 200-213.
Hodgkinson, A., and Eyre-Walker, A. (2011). Variation in the mutation rate
across mammalian genomes. Nat Rev Genet 12, 756-766.
Hodis, E., Watson, I.R., Kryukov, G.V., Arold, S.T., Imielinski, M., Theurillat, J.P.,
Nickerson, E., Auclair, D., Li, L., Place, C., et al. (2012). A landscape of driver
mutations in melanoma. Cell 150, 251-263.
Holmfeldt, L., Wei, L., Diaz-Flores, E., Walsh, M., Zhang, J., Ding, L., Payne-
Turner, D., Churchman, M., Andersson, A., Chen, S.C., et al. (2013). The
genomic landscape of hypodiploid acute lymphoblastic leukemia. Nat Genet.
Hon, G.C., Hawkins, R.D., Caballero, O.L., Lo, C., Lister, R., Pelizzola, M.,
Valsesia, A., Ye, Z., Kuan, S., Edsall, L.E., et al. (2012). Global DNA
hypomethylation coupled to repressive chromatin domain formation and gene
silencing in breast cancer. Genome Res 22, 246-258.
42
Hussin, J., Sinnett, D., Casals, F., Idaghdour, Y., Bruat, V., Saillour, V., Healy, J.,
Grenier, J.C., de Malliard, T., Busche, S., et al. (2013). Rare allelic forms of
PRDM9 associated with childhood leukemogenesis. Genome Res.
Hyland, E.M., Molina, H., Poorey, K., Jie, C., Xie, Z., Dai, J., Qian, J., Bekiranov,
S., Auble, D.T., Pandey, A., et al. (2011). An evolutionarily 'young' lysine residue
in histone H3 attenuates transcriptional output in Saccharomyces cerevisiae.
Genes & development 25, 1306-1319.
Imielinski, M., Berger, A.H., Hammerman, P.S., Hernandez, B., Pugh, T.J., Hodis,
E., Cho, J., Suh, J., Capelletti, M., Sivachenko, A., et al. (2012). Mapping the
hallmarks of lung adenocarcinoma with massively parallel sequencing. Cell 150,
1107-1120.
Inthal, A., Zeitlhofer, P., Zeginigg, M., Morak, M., Grausenburger, R., Fronkova,
E., Fahrner, B., Mann, G., Haas, O.A., and Panzer-Grumayer, R. (2012).
CREBBP HAT domain mutations prevail in relapse cases of high hyperdiploid
childhood acute lymphoblastic leukemia. Leukemia 26, 1797-1803.
Irizarry, R.A., Ladd-Acosta, C., Wen, B., Wu, Z., Montano, C., Onyango, P., Cui,
H., Gabo, K., Rongione, M., Webster, M., et al. (2009). The human colon cancer
methylome shows similar hypo- and hypermethylation at conserved tissue-
specific CpG island shores. Nat Genet 41, 178-186.
Ito, S., Shen, L., Dai, Q., Wu, S.C., Collins, L.B., Swenberg, J.A., He, C., and
Zhang, Y. (2011). Tet proteins can convert 5-methylcytosine to 5-formylcytosine
and 5-carboxylcytosine. Science 333, 1300-1303.
Jiao, Y., Shi, C., Edil, B.H., de Wilde, R.F., Klimstra, D.S., Maitra, A., Schulick,
R.D., Tang, L.H., Wolfgang, C.L., Choti, M.A., et al. (2011). DAXX/ATRX, MEN1,
and mTOR pathway genes are frequently altered in pancreatic neuroendocrine
tumors. Science 331, 1199-1203.
Jin, C., Zang, C., Wei, G., Cui, K., Peng, W., Zhao, K., and Felsenfeld, G. (2009).
H3.3/H2A.Z double variant-containing nucleosomes mark 'nucleosome-free
regions' of active promoters and other regulatory regions. Nat Genet 41, 941-945.
Johnson, D.G., and Dent, S.Y.R. (2013). Chromatin: Receiver and Quarterback
for Cellular Signals. Cell 152, In Press.
Jones, D.T., Jager, N., Kool, M., Zichner, T., Hutter, B., Sultan, M., Cho, Y.J.,
Pugh, T.J., Hovestadt, V., Stutz, A.M., et al. (2012). Dissecting the genomic
complexity underlying medulloblastoma. Nature 488, 100-105.
43
Jones, P.A. (2012). Functions of DNA methylation: islands, start sites, gene
bodies and beyond. Nat Rev Genet 13, 484-492.
Jones, P.A., and Baylin, S.B. (2002). The fundamental role of epigenetic events
in cancer. Nat Rev Genet 3, 415-428.
Jones, P.A., and Laird, P.W. (1999). Cancer epigenetics comes of age. Nature
Genetics 21, 163-167.
Jones, P.A., and Liang, G. (2009). Rethinking how DNA methylation patterns are
maintained. Nat Rev Genet 10, 805-811.
Jones, S., Wang, T.L., Shih Ie, M., Mao, T.L., Nakayama, K., Roden, R., Glas, R.,
Slamon, D., Diaz, L.A., Jr., Vogelstein, B., et al. (2010). Frequent mutations of
chromatin remodeling gene ARID1A in ovarian clear cell carcinoma. Science 330,
228-231.
Kamakaka, R.T., and Biggins, S. (2005). Histone variants: deviants? Genes Dev
19, 295-310.
Kerkel, K., Spadola, A., Yuan, E., Kosek, J., Jiang, L., Hod, E., Li, K., Murty, V.V.,
Schupf, N., Vilain, E., et al. (2008). Genomic surveys by methylation-sensitive
SNP analysis identify sequence- dependent allele-specific DNA methylation. Nat
Genet 40, 904-908.
Kim, M., Trinh, B.N., Long, T.I., Oghamian, S., and Laird, P.W. (2004). Dnmt1
deficiency leads to enhanced microsatellite instability in mouse embryonic stem
cells. Nucleic acids research 32, 5742-5749.
Kim, T.H., Abdullaev, Z.K., Smith, A.D., Ching, K.A., Loukinov, D.I., Green, R.D.,
Zhang, M.Q., Lobanenkov, V.V., and Ren, B. (2007). Analysis of the vertebrate
insulator protein CTCF-binding sites in the human genome. Cell 128, 1231-1245.
Ko, M., Huang, Y., Jankowska, A.M., Pape, U.J., Tahiliani, M., Bandukwala, H.S.,
An, J., Lamperti, E.D., Koh, K.P., Ganetzky, R., et al. (2010). Impaired
hydroxylation of 5-methylcytosine in myeloid cancers with mutant TET2. Nature
468, 839-843.
Kondo, Y., Shen, L., Cheng, A.S., Ahmed, S., Boumber, Y., Charo, C., Yamochi,
T., Urano, T., Furukawa, K., Kwabi-Addo, B., et al. (2008). Gene silencing in
cancer by histone H3 lysine 27 trimethylation independent of promoter DNA
methylation. Nat Genet 40, 741-750.
44
Ku, M., Koche, R.P., Rheinbay, E., Mendenhall, E.M., Endoh, M., Mikkelsen, T.S.,
Presser, A., Nusbaum, C., Xie, X., Chi, A.S., et al. (2008). Genomewide analysis
of PRC1 and PRC2 occupancy identifies two classes of bivalent domains. PLoS
Genet 4, E1000242.
Lai, A.Y., and Wade, P.A. (2011). Cancer biology and NuRD: a multifaceted
chromatin remodelling complex. Nature reviews Cancer 11, 588-596.
Laird, P.W. (2010). Principles and challenges of genomewide DNA methylation
analysis. Nature reviews Genetics 11, 191-203.
Laird, P.W., Jackson-Grusby, L., Fazeli, A., Dickinson, S.L., Jung, W.E., Li, E.,
Weinberg, R.A., and Jaenisch, R. (1995). Suppression of intestinal neoplasia by
DNA hypomethylation. Cell 81, 197-205.
Langemeijer, S.M., Kuiper, R.P., Berends, M., Knops, R., Aslanyan, M.G.,
Massop, M., Stevens-Linders, E., van Hoogen, P., van Kessel, A.G., Raymakers,
R.A., et al. (2009). Acquired mutations in TET2 are common in myelodysplastic
syndromes. Nat Genet 41, 838-842.
Le Gallo, M., O'Hara, A.J., Rudd, M.L., Urick, M.E., Hansen, N.F., O'Neil, N.J.,
Price, J.C., Zhang, S., England, B.M., Godwin, A.K., et al. (2012). Exome
sequencing of serous endometrial tumors identifies recurrent somatic mutations
in chromatin-remodeling and ubiquitin ligase complex genes. Nat Genet 44,
1310-1315.
Lee, E., Iskow, R., Yang, L., Gokcumen, O., Haseley, P., Luquette, L.J., 3rd, Lohr,
J.G., Harris, C.C., Ding, L., Wilson, R.K., et al. (2012). Landscape of somatic
retrotransposition in human cancers. Science 337, 967-971.
Lee, J.T. (2012). Epigenetic regulation by long noncoding RNAs. Science 338,
1435-1439.
Ley, T.J., Ding, L., Walter, M.J., McLellan, M.D., Lamprecht, T., Larson, D.E.,
Kandoth, C., Payton, J.E., Baty, J., Welch, J., et al. (2010). DNMT3A mutations
in acute myeloid leukemia. The New England journal of medicine 363, 2424-2433.
Li, J., Harris, R.A., Cheung, S.W., Coarfa, C., Jeong, M., Goodell, M.A., White,
L.D., Patel, A., Kang, S.H., Shaw, C., et al. (2012). Genomic hypomethylation in
the human germline associates with selective structural mutability in the human
genome. PLoS Genet 8, e1002692.
45
Lienert, F., Wirbelauer, C., Som, I., Dean, A., Mohn, F., and Schubeler, D. (2011).
Identification of genetic elements that autonomously determine DNA methylation
states. Nature Genetics 43, 1091-1097.
Ligtenberg, M.J., Kuiper, R.P., Chan, T.L., Goossens, M., Hebeda, K.M.,
Voorendt, M., Lee, T.Y., Bodmer, D., Hoenselaar, E., Hendriks-Cornelissen, S.J.,
et al. (2009). Heritable somatic methylation and inactivation of MSH2 in families
with Lynch syndrome due to deletion of the 3' exons of TACSTD1. Nat Genet 41,
112-117.
Lister, R., Pelizzola, M., Dowen, R.H., Hawkins, R.D., Hon, G., Tonti-Filippini, J.,
Nery, J.R., Lee, L., Ye, Z., Ngo, Q.M., et al. (2009). Human DNA methylomes at
base resolution show widespread epigenomic differences. Nature 462, 315-322.
Liu, G., Bollig-Fischer, A., Kreike, B., van de Vijver, M.J., Abrams, J., Ethier, S.P.,
and Yang, Z.Q. (2009). Genomic amplification and oncogenic properties of the
GASC1 histone demethylase gene in breast cancer. Oncogene 28, 4491-4500.
Lohr, J.G., Stojanov, P., Lawrence, M.S., Auclair, D., Chapuy, B., Sougnez, C.,
Cruz-Gordillo, P., Knoechel, B., Asmann, Y.W., Slager, S.L., et al. (2012).
Discovery and prioritization of somatic mutations in diffuse large B-cell lymphoma
(DLBCL) by whole-exome sequencing. Proc Natl Acad Sci U S A 109, 3879-3884.
Loughery, J.E., Dunne, P.D., O'Neill, K.M., Meehan, R.R., McDaid, J.R., and
Walsh, C.P. (2011). DNMT1 deficiency triggers mismatch repair defects in
human cells through depletion of repair protein levels in a process involving the
DNA damage response. Human molecular genetics 20, 3241-3255.
Love, C., Sun, Z., Jima, D., Li, G., Zhang, J., Miles, R., Richards, K.L., Dunphy,
C.H., Choi, W.W., Srivastava, G., et al. (2012). The genetic landscape of
mutations in Burkitt lymphoma. Nat Genet 44, 1321-1325.
Lu, C., Ward, P.S., Kapoor, G.S., Rohle, D., Turcan, S., Abdel-Wahab, O.,
Edwards, C.R., Khanin, R., Figueroa, M.E., Melnick, A., et al. (2012). IDH
mutation impairs histone demethylation and results in a block to cell
differentiation. Nature 483, 474-478.
Meuleman, W., Peric-Hupkes, D., Kind, J., Beaudry, J.B., Pagie, L., Kellis, M.,
Reinders, M., Wessels, L., and van Steensel, B. (2012). Constitutive nuclear
lamina-genome interactions are highly conserved and associated with A/T-rich
sequence. Genome Res.
46
Mills, A.A. (2010). Throwing the cancer switch: reciprocal roles of polycomb and
trithorax proteins. Nat Rev Cancer 10, 669-682.
Minucci, S., and Pelicci, P.G. (2006). Histone deacetylase inhibitors and the
promise of epigenetic (and more) treatments for cancer. Nat Rev Cancer 6, 38-
51.
Mohn, F., Weber, M., Rebhan, M., Roloff, T.C., Richter, J., Stadler, M.B., Bibel,
M., and Schubeler, D. (2008). Lineage-specific polycomb targets and de novo
DNA methylation define restriction and potential of neuronal progenitors. Mol Cell
30, 755-766.
Morin, R.D., Johnson, N.A., Severson, T.M., Mungall, A.J., An, J., Goya, R., Paul,
J.E., Boyle, M., Woolcock, B.W., Kuchenbauer, F., et al. (2010). Somatic
mutations altering EZH2 (Tyr641) in follicular and diffuse large B-cell lymphomas
of germinal-center origin. Nat Genet 42, 181-185.
Morin, R.D., Mendez-Lago, M., Mungall, A.J., Goya, R., Mungall, K.L., Corbett,
R.D., Johnson, N.A., Severson, T.M., Chiu, R., Field, M., et al. (2011). Frequent
mutation of histone-modifying genes in non-Hodgkin lymphoma. Nature 476,
298-303.
Mortusewicz, O., Schermelleh, L., Walter, J., Cardoso, M.C., and Leonhardt, H.
(2005). Recruitment of DNA methyltransferase I to DNA repair sites. Proc Natl
Acad Sci U S A 102, 8905-8909.
Mullighan, C.G., Zhang, J., Kasper, L.H., Lerach, S., Payne-Turner, D., Phillips,
L.A., Heatley, S.L., Holmfeldt, L., Collins-Underwood, J.R., Ma, J., et al. (2011).
CREBBP mutations in relapsed acute lymphoblastic leukaemia. Nature 471, 235-
239.
Mummert, S.K., Lobanenkov, V.A., and Feinberg, A.P. (2005). Association of
chromosome arm 16q loss with loss of imprinting of insulin-like growth factor-II in
Wilms tumor. Genes Chromosomes Cancer 43, 155-161.
Musselman, C.A., Lalonde, M.E., Cote, J., and Kutateladze, T.G. (2012).
Perceiving the epigenetic landscape through histone readers. Nat Struct Mol Biol
19, 1218-1227.
Nikoloski, G., Langemeijer, S.M., Kuiper, R.P., Knops, R., Massop, M.,
Tonnissen, E.R., van der Heijden, A., Scheele, T.N., Vandenberghe, P., de Witte,
T., et al. (2010). Somatic mutations of the histone methyltransferase gene EZH2
in myelodysplastic syndromes. Nat Genet 42, 665-667.
47
Noushmehr, H., Weisenberger, D.J., Diefes, K., Phillips, H.S., Pujara, K., Berman,
B.P., Pan, F., Pelloski, C.E., Sulman, E.P., Bhat, K.P., et al. (2010). Identification
of a CpG island methylator phenotype that defines a distinct subgroup of glioma.
Cancer Cell 17, 510-522.
Ntziachristos, P., Tsirigos, A., Van Vlierberghe, P., Nedjic, J., Trimarchi, T.,
Flaherty, M.S., Ferres-Marco, D., da Ros, V., Tang, Z., Siegle, J., et al. (2012).
Genetic inactivation of the polycomb repressive complex 2 in T cell acute
lymphoblastic leukemia. Nat Med 18, 298-301.
Ohm, J.E., McGarvey, K.M., Yu, X., Cheng, L., Schuebel, K.E., Cope, L.,
Mohammad, H.P., Chen, W., Daniel, V.C., Yu, W., et al. (2007). A stem cell-like
chromatin pattern may predispose tumor suppressor genes to DNA
hypermethylation and heritable silencing. Nat Genet 39, 237-242.
Okano, M., Bell, D.W., Haber, D.A., and Li, E. (1999). DNA methyltransferases
Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian
development. Cell 99, 247-257.
Olivier, M., Hollstein, M., and Hainaut, P. (2010). TP53 mutations in human
cancers: origins, consequences, and clinical use. Cold Spring Harb Perspect Biol
2, a001008.
Papamichos-Chronakis, M., and Peterson, C.L. (2012). Chromatin and the
genome integrity network. Nat Rev Genet 14, 62-75.
Parsons, D.W., Li, M., Zhang, X., Jones, S., Leary, R.J., Lin, J.C., Boca, S.M.,
Carter, H., Samayoa, J., Bettegowda, C., et al. (2011). The genetic landscape of
the childhood cancer medulloblastoma. Science 331, 435-439.
Pasqualucci, L., Trifonov, V., Fabbri, G., Ma, J., Rossi, D., Chiarenza, A., Wells,
V.A., Grunn, A., Messina, M., Elliot, O., et al. (2011). Analysis of the coding
genome of diffuse large B-cell lymphoma. Nat Genet 43, 830-837.
Petruk, S., Sedkov, Y., Johnston, D.M., Hodgson, J.W., Black, K.L., Kovermann,
S.K., Beck, S., Canaani, E., Brock, H.W., and Mazo, A. (2012). TrxG and PcG
proteins but not methylated histones remain associated with DNA through
replication. Cell 150, 922-933.
Prokhortchouk, A., Sansom, O., Selfridge, J., Caballero, I.M., Salozhin, S.,
Aithozhina, D., Cerchietti, L., Meng, F.G., Augenlicht, L.H., Mariadason, J.M., et
al. (2006). Kaiso-deficient mice show resistance to intestinal cancer. Mol Cell Biol
26, 199-208.
48
Pugh, T.J., Weeraratne, S.D., Archer, T.C., Pomeranz Krummel, D.A., Auclair, D.,
Bochicchio, J., Carneiro, M.O., Carter, S.L., Cibulskis, K., Erlich, R.L., et al.
(2012). Medulloblastoma exome sequencing uncovers subtype-specific somatic
mutations. Nature 488, 106-110.
Quivoron, C., Couronne, L., Della Valle, V., Lopez, C.K., Plo, I., Wagner-Ballon,
O., Do Cruzeiro, M., Delhommeau, F., Arnulf, B., Stern, M.H., et al. (2011). TET2
inactivation results in pleiotropic hematopoietic abnormalities in mouse and is a
recurrent event during human lymphomagenesis. Cancer Cell 20, 25-38.
Raddatz, G., Gao, Q., Bender, S., Jaenisch, R., and Lyko, F. (2012). Dnmt3a
Protects Active Chromosome Domains against Cancer-Associated
Hypomethylation. PLoS Genet 8, e1003146.
Ram, O., Goren, A., Amit, I., Shoresh, N., Yosef, N., Ernst, J., Kellis, M., Gymrek,
M., Issner, R., Coyne, M., et al. (2011). Combinatorial patterning of chromatin
regulators uncovered by genome-wide location analysis in human cells. Cell 147,
1628-1639.
Robinson, G., Parker, M., Kranenburg, T.A., Lu, C., Chen, X., Ding, L., Phoenix,
T.N., Hedlund, E., Wei, L., Zhu, X., et al. (2012). Novel mutations target distinct
subgroups of medulloblastoma. Nature 488, 43-48.
Rui, L., Emre, N.C., Kruhlak, M.J., Chung, H.J., Steidl, C., Slack, G., Wright,
G.W., Lenz, G., Ngo, V.N., Shaffer, A.L., et al. (2010). Cooperative epigenetic
modulation by cancer amplicon genes. Cancer Cell 18, 590-605.
Sandoval, J., and Esteller, M. (2012). Cancer epigenomics: beyond genomics.
Curr Opin Genet Dev 22, 50-55.
Sansom, O.J., Berger, J., Bishop, S.M., Hendrich, B., Bird, A., and Clarke, A.R.
(2003). Deficiency of Mbd2 suppresses intestinal tumorigenesis. Nat Genet 34,
145-147.
Schlesinger, Y., Straussman, R., Keshet, I., Farkash, S., Hecht, M., Zimmerman,
J., Eden, E., Yakhini, Z., Ben-Shushan, E., Reubinoff, B.E., et al. (2007).
Polycomb-mediated methylation on Lys27 of histone H3 pre-marks genes for de
novo methylation in cancer. Nat Genet 39, 232-236.
Schneppenheim, R., Fruhwald, M.C., Gesk, S., Hasselblatt, M., Jeibmann, A.,
Kordes, U., Kreuz, M., Leuschner, I., Martin Subero, J.I., Obser, T., et al. (2010).
Germline nonsense mutation and somatic inactivation of SMARCA4/BRG1 in a
49
family with rhabdoid tumor predisposition syndrome. Am J Hum Genet 86, 279-
284.
Schuster-Bockler, B., and Lehner, B. (2012). Chromatin organization is a major
influence on regional mutation rates in human cancer cells. Nature 488, 504-507.
Schwartzentruber, J., Korshunov, A., Liu, X.Y., Jones, D.T., Pfaff, E., Jacob, K.,
Sturm, D., Fontebasso, A.M., Quang, D.A., Tonjes, M., et al. (2012). Driver
mutations in histone H3.3 and chromatin remodelling genes in paediatric
glioblastoma. Nature 482, 226-231.
Seligson, D.B., Horvath, S., Shi, T., Yu, H., Tze, S., Grunstein, M., and Kurdistani,
S.K. (2005). Global histone modification patterns predict risk of prostate cancer
recurrence. Nature 435, 1262-1266.
Sharma, S., De Carvalho, D.D., Jeong, S., Jones, P.A., and Liang, G. (2011).
Nucleosomes containing methylated DNA stabilize DNA methyltransferases
3A/3B and ensure faithful epigenetic inheritance. PLoS genetics 7, e1001286.
Simon, C., Chagraoui, J., Krosl, J., Gendron, P., Wilhelm, B., Lemieux, S.,
Boucher, G., Chagnon, P., Drouin, S., Lambert, R., et al. (2012). A key role for
EZH2 and associated genes in mouse and human adult T-cell acute leukemia.
Genes Dev 26, 651-656.
Simpkins, S.B., Bocker, T., Swisher, E.M., Mutch, D.G., Gersell, D.J., Kovatich,
A.J., Palazzo, J.P., Fishel, R., and Goodfellow, P.J. (1999). MLH1 promoter
methylation and gene silencing is the primary cause of microsatellite instability in
sporadic endometrial cancers. Hum Mol Genet 8, 661-666.
Smith, M.J., O'Sullivan, J., Bhaskar, S.S., Hadfield, K.D., Poke, G., Caird, J.,
Sharif, S., Eccles, D., Fitzpatrick, D., Rawluk, D., et al. (2013). Loss-of-function
mutations in SMARCE1 cause an inherited disorder of multiple spinal
meningiomas. Nat Genet.
Sproul, D., Kitchen, R.R., Nestor, C.E., Dixon, J.M., Sims, A.H., Harrison, D.J.,
Ramsahoye, B.H., and Meehan, R.R. (2012). Tissue of origin determines cancer-
associated CpG island promoter hypermethylation patterns. Genome Biol 13,
R84.
Stransky, N., Egloff, A.M., Tward, A.D., Kostic, A.D., Cibulskis, K., Sivachenko,
A., Kryukov, G.V., Lawrence, M.S., Sougnez, C., McKenna, A., et al. (2011). The
mutational landscape of head and neck squamous cell carcinoma. Science 333,
1157-1160.
50
Sturm, D., Witt, H., Hovestadt, V., Khuong-Quang, D.A., Jones, D.T., Konermann,
C., Pfaff, E., Tonjes, M., Sill, M., Bender, S., et al. (2012). Hotspot mutations in
H3F3A and IDH1 define distinct epigenetic and biological subgroups of
glioblastoma. Cancer Cell 22, 425-437.
Tan, J., Jones, M., Koseki, H., Nakayama, M., Muntean, A.G., Maillard, I., and
Hess, J.L. (2011a). CBX8, a polycomb group protein, is essential for MLL-AF9-
induced leukemogenesis. Cancer Cell 20, 563-575.
Tan, M., Luo, H., Lee, S., Jin, F., Yang, J.S., Montellier, E., Buchou, T., Cheng,
Z., Rousseaux, S., Rajagopal, N., et al. (2011b). Identification of 67 histone
marks and histone lysine crotonylation as a new type of histone modification. Cell
146, 1016-1028.
Tanay, A., O'Donnell, A.H., Damelin, M., and Bestor, T.H. (2007).
Hyperconserved CpG domains underlie Polycomb-binding sites. Proc Natl Acad
Sci U S A 104, 5521-5526.
TCGA, C.G.A.R.N. (2011). Integrated genomic analyses of ovarian carcinoma.
Nature 474, 609-615.
TCGA, C.G.A.R.N. (2012a). Comprehensive genomic characterization of
squamous cell lung cancers. Nature 489, 519-525.
TCGA, C.G.A.R.N. (2012b). Comprehensive molecular characterization of
human colon and rectal cancer. Nature 487, 330-337.
TCGA, C.G.A.R.N. (2012c). Comprehensive molecular portraits of human breast
tumours. Nature 490, 61-70.
Teschendorff, A.E., Menon, U., Gentry-Maharaj, A., Ramus, S.J., Weisenberger,
D.J., Shen, H., Campan, M., Noushmehr, H., Bell, C.G., Maxwell, A.P., et al.
(2010). Age-dependent DNA methylation of genes that are suppressed in stem
cells is a hallmark of cancer. Genome research 20, 440-446.
Testa, J.R., Cheung, M., Pei, J., Below, J.E., Tan, Y., Sementino, E., Cox, N.J.,
Dogan, A.U., Pass, H.I., Trusa, S., et al. (2011). Germline BAP1 mutations
predispose to malignant mesothelioma. Nat Genet 43, 1022-1025.
Toyota, M., Ahuja, N., Ohe-Toyota, M., Herman, J.G., Baylin, S.B., and Issa, J.P.
(1999). CpG island methylator phenotype in colorectal cancer. Proc Natl Acad
Sci U S A 96, 8681-8686.
51
Toyota, M., and Suzuki, H. (2010). Epigenetic drivers of genetic alterations. Adv
Genet 70, 309-323.
Turcan, S., Rohle, D., Goenka, A., Walsh, L.A., Fang, F., Yilmaz, E., Campos, C.,
Fabius, A.W., Lu, C., Ward, P.S., et al. (2012). IDH1 mutation is sufficient to
establish the glioma hypermethylator phenotype. Nature 483, 479-483.
Valouev, A., Johnson, S.M., Boyd, S.D., Smith, C.L., Fire, A.Z., and Sidow, A.
(2011). Determinants of nucleosome organization in primary human cells. Nature
474, 516-520.
van Haaften, G., Dalgliesh, G.L., Davies, H., Chen, L., Bignell, G., Greenman, C.,
Edkins, S., Hardy, C., O'Meara, S., Teague, J., et al. (2009). Somatic mutations
of the histone H3K27 demethylase gene UTX in human cancer. Nat Genet 41,
521-523.
Van Vlierberghe, P., Palomero, T., Khiabanian, H., Van der Meulen, J., Castillo,
M., Van Roy, N., De Moerloose, B., Philippe, J., Gonzalez-Garcia, S., Toribio,
M.L., et al. (2010). PHF6 mutations in T-cell acute lymphoblastic leukemia. Nat
Genet 42, 338-342.
Van Vlierberghe, P., Patel, J., Abdel-Wahab, O., Lobry, C., Hedvat, C.V., Balbin,
M., Nicolas, C., Payer, A.R., Fernandez, H.F., Tallman, M.S., et al. (2011). PHF6
mutations in adult acute myeloid leukemia. Leukemia 25, 130-134.
Varambally, S., Cao, Q., Mani, R.S., Shankar, S., Wang, X., Ateeq, B., Laxman,
B., Cao, X., Jing, X., Ramnarayanan, K., et al. (2008). Genomic loss of
microRNA-101 leads to overexpression of histone methyltransferase EZH2 in
cancer. Science 322, 1695-1699.
Varela, I., Tarpey, P., Raine, K., Huang, D., Ong, C.K., Stephens, P., Davies, H.,
Jones, D., Lin, M.L., Teague, J., et al. (2011). Exome sequencing identifies
frequent mutation of the SWI/SNF complex gene PBRM1 in renal carcinoma.
Nature 469, 539-542.
Versteege, I., Sevenet, N., Lange, J., Rousseau-Merck, M.F., Ambros, P.,
Handgretinger, R., Aurias, A., and Delattre, O. (1998). Truncating mutations of
hSNF5/INI1 in aggressive paediatric cancer. Nature 394, 203-206.
Wang, R.Y., Kuo, K.C., Gehrke, C.W., Huang, L.H., and Ehrlich, M. (1982). Heat-
and alkali-induced deamination of 5-methylcytosine and cytosine residues in
DNA. Biochim Biophys Acta 697, 371-377.
52
Wang, Z.Y., and Chen, Z. (2008). Acute promyelocytic leukemia: from highly fatal
to highly curable. Blood 111, 2505-2515.
Weisenberger, D.J., Siegmund, K.D., Campan, M., Young, J., Long, T.I., Faasse,
M.A., Kang, G.H., Widschwendter, M., Weener, D., Buchanan, D., et al. (2006).
CpG island methylator phenotype underlies sporadic microsatellite instability and
is tightly associated with BRAF mutation in colorectal cancer. Nature Genetics 38,
787-793.
Weissmann, S., Alpermann, T., Grossmann, V., Kowarsch, A., Nadarajah, N.,
Eder, C., Dicker, F., Fasan, A., Haferlach, C., Haferlach, T., et al. (2012).
Landscape of TET2 mutations in acute myeloid leukemia. Leukemia 26, 934-942.
Widschwendter, M., Fiegl, H., Egle, D., Mueller-Holzner, E., Spizzo, G., Marth, C.,
Weisenberger, D.J., Campan, M., Young, J., Jacobs, I., et al. (2007). Epigenetic
stem cell signature in cancer. Nature Genetics 39, 157-158.
Wiegand, K.C., Shah, S.P., Al-Agha, O.M., Zhao, Y., Tse, K., Zeng, T., Senz, J.,
McConechy, M.K., Anglesio, M.S., Kalloger, S.E., et al. (2010). ARID1A
mutations in endometriosis-associated ovarian carcinomas. N Engl J Med 363,
1532-1543.
Wiesner, T., Obenauf, A.C., Murali, R., Fried, I., Griewank, K.G., Ulz, P.,
Windpassinger, C., Wackernagel, W., Loy, S., Wolf, I., et al. (2011). Germline
mutations in BAP1 predispose to melanocytic tumors. Nat Genet 43, 1018-1021.
Williams, K., Christensen, J., and Helin, K. (2012). DNA methylation: TET
proteins-guardians of CpG islands? EMBO Rep 13, 28-35.
Wilson, B.G., and Roberts, C.W. (2011). SWI/SNF nucleosome remodellers and
cancer. Nat Rev Cancer 11, 481-492.
Wu, G., Broniscer, A., McEachron, T.A., Lu, C., Paugh, B.S., Becksfort, J., Qu,
C., Ding, L., Huether, R., Parker, M., et al. (2012). Somatic histone H3 alterations
in pediatric diffuse intrinsic pontine gliomas and non-brainstem glioblastomas.
Nat Genet 44, 251-253.
Xu, K., Wu, Z.J., Groner, A.C., He, H.H., Cai, C., Lis, R.T., Wu, X., Stack, E.C.,
Loda, M., Liu, T., et al. (2012). EZH2 oncogenic activity in castration-resistant
prostate cancer cells is Polycomb-independent. Science 338, 1465-1469.
53
Xu, W., Yang, H., Liu, Y., Yang, Y., Wang, P., Kim, S.H., Ito, S., Yang, C., Xiao,
M.T., Liu, L.X., et al. (2011). Oncometabolite 2-hydroxyglutarate is a competitive
inhibitor of alpha-ketoglutarate-dependent dioxygenases. Cancer Cell 19, 17-30.
Yamamoto, E., Suzuki, H., Yamano, H.O., Maruyama, R., Nojima, M., Kamimae,
S., Sawada, T., Ashida, M., Yoshikawa, K., Kimura, T., et al. (2012). Molecular
dissection of premalignant colorectal lesions reveals early onset of the CpG
island methylator phenotype. Am J Pathol 181, 1847-1861.
Yamashita, K., Dai, T., Dai, Y., Yamamoto, F., and Perucho, M. (2003). Genetics
supersedes epigenetics in colon cancer phenotype. Cancer Cell 4, 121-131.
Yamazaki, J., Estecio, M.R., Lu, Y., Long, H., Malouf, G.G., Graber, D., Huo, Y.,
Ramagli, L., Liang, S., Kornblau, S.M., et al. (2013). The epigenome of AML
stem and progenitor cells. Epigenetics 8, 92-104.
Yan, X.J., Xu, J., Gu, Z.H., Pan, C.M., Lu, G., Shen, Y., Shi, J.Y., Zhu, Y.M.,
Tang, L., Zhang, X.W., et al. (2011). Exome sequencing identifies somatic
mutations of DNA methyltransferase gene DNMT3A in acute monocytic leukemia.
Nature genetics 43, 309-315.
Yang, X.J. (2004). The diverse superfamily of lysine acetyltransferases and their
roles in leukemia and other diseases. Nucleic Acids Res 32, 959-976.
Yokoyama, A., and Cleary, M.L. (2008). Menin critically links MLL proteins with
LEDGF on cancer-associated target genes. Cancer Cell 14, 36-46.
You, J.S., and Jones, P.A. (2012). Cancer genetics and epigenetics: two sides of
the same coin? Cancer Cell 22, 9-20.
Zeisig, B.B., Kulasekararaj, A.G., Mufti, G.J., and So, C.W. (2012). SnapShot:
Acute myeloid leukemia. Cancer Cell 22, 698-698 e691.
Zhang, J., Ding, L., Holmfeldt, L., Wu, G., Heatley, S.L., Payne-Turner, D.,
Easton, J., Chen, X., Wang, J., Rusch, M., et al. (2012). The genetic basis of
early T-cell precursor acute lymphoblastic leukaemia. Nature 481, 157-163.
54
Zhu, J., Adli, M., Zou, J.Y., Verstappen, G., Coyne, M., Zhang, X., Durham, T.,
Miri, M., Deshpande, V., De Jager, P.L., et al. (2013). Genome-wide Chromatin
State Transitions Associated with Developmental and Environmental Cues. Cell
152, 642-654.
Zullo, J.M., Demarco, I.A., Pique-Regi, R., Gaffney, D.J., Epstein, C.B., Spooner,
C.J., Luperchio, T.R., Bernstein, B.E., Pritchard, J.K., Reddy, K.L., et al. (2012).
DNA sequence-dependent compartmentalization and silencing of chromatin at
the nuclear lamina. Cell 149, 1474-1487.
55
Chapter Two: Epithelial Ovarian Cancer
Introduction
Epithelial ovarian carcinoma is the fifth leading cause of cancer death and
the leading cause of death from gynecological cancer in women in the US, with a
five-year survival rate of about 44% (American Cancer Society, 2013). Currently,
the histological subtypes are the most important guidelines in instructing clinical
treatment. There are four major histological types of epithelial ovarian cancer,
namely serous, endometrioid, clear cell and mucinous carcinomas. Kurman and
colleagues have proposed a new classification scheme in which two types of
ovarian cancers are described (Kurman and Shih Ie, 2010). The clinically indolent
type I cancers include low-grade serous, low-grade endometrioid, clear cell and
mucinous, and the more aggressive type II cancers include high-grade serous and
high-grade endometrioid tumors. However, this classification scheme does not
suffice to explain the difference in epidemiological risk factors or clinical outcome,
and thus complementary or even more informative classification is needed. Our
preliminary studies also show that histological subtype does not have a deterministic
effect on the DNA methylation profile. What is more, the lack of efficient early
detection methods and successful treatment strategies for ovarian cancer entails
more efforts to identify molecular abnormalities contributing to the disease.
The majority of deaths result from advanced stage, high grade serous
ovarian cancer, which represents about 70% of all ovarian cancers) (Seidman et al.,
2004). Aggressive surgery followed by adjuvant taxane and platinum-based
combination chemotherapy remains the standard of care for patients with newly
diagnosed advanced ovarian carcinoma. However, approximately 25% of patients
56
receiving platinum-based therapy will recur within 6 months (defined as platinum-
resistant disease)(Miller et al., 2009), and most of the rest will recur over the next
few years. The 5-year overall survival for advanced ovarian cancer is only 27%
(American Cancer Society, 2013). Therefore, most of our effort will be put into high
grade, high stage serous ovarian cystadenocarcinomas, aiming at better
understanding the disease heterogeneity and its associated etiology, and thereby
hopefully improving the current treatment paradigm as well as prognosis.
57
Results
Cancer-associated epigenetic silencing of genes
Epigenetic silencing is increasingly recognized as an important alternative to
deletion or mutation in the inactivation of gene function in cancer(Jones and Baylin,
2002; Jones and Laird, 1999). The identification of genes that are epigenetically
silenced in a cancer-specific manner adds to our understanding of the full
complement of molecular alterations that contribute to oncogenesis, and might shed
light on the early detection, prevention, treatment and prognosis of the disease.
DNA methylation, repressive histone modifications, and other marks can
work in concert to achieve an aberrant epigenetically silenced state at susceptible
gene promoters. For the purposes of this study, the assessment of epigenetic marks
is confined to the quantitative measurement of DNA methylation levels at a limited
number of CpG dinucleotides (an average of two per promoter) at 14,475 genes,
using the Illumina Infinium HumanMethylation27 BeadChip assay (see Methods).
The goal of this section is to identify genes with evidence for cancer-specific
promoter hypermethylation with an associated decrease in gene expression. The
general properties of such genes are: low levels of promoter methylation in control
tissues thought to represent potential cells of origin, high levels of promoter
methylation and an associated lower expression in at least some of the tumors, and
a good inverse correlation between promoter methylation levels and gene
expression. The relationship between DNA methylation and gene expression is
complex and highly variable among the 12,233 genes in our data set for which we
have both DNA methylation and gene expression measurements. Therefore, the
58
appropriate selection criteria may vary depending on the nature of this relationship.
In designing a strategy to identify epigenetically silenced genes, we considered the
following three issues.
First, the impact of CpG methylation on transcriptional potential depends on
the density of the methylated CpGs, and their location relative to the transcription
start site and functional promoter elements. The constraints of the
HumanMethylation27 BeadChip design, with an average of two CpGs per promoter,
do not allow for a comprehensive assessment of each gene, and the locations of the
measured CpGs may be uninformative for some genes. Therefore, we anticipate
that we will be unable to identify some silenced genes. Some genes may have
alternative promoters for which the methylation status is not assessed. Genes
lacking promoter methylation in some tumors or in normal tissues may not be
expressed due to a lack of appropriate transcription factors. These factors all
contribute to a complex relationship between our DNA methylation measurements
and observed gene expression levels. We used a rank-based Spearman correlation
to allow for nonlinear relationships between DNA methylation and gene expression.
Second, the identification of cancer-associated DNA methylation alterations
requires a comparison of tumor DNA methylation data to a control tissue, ideally
representing (or at least enriched for) the cell-of-origin of serous ovarian cancer.
The normal ovary surface epithelium and fallopian tube epithelium have both been
proposed as originating tissues for ovarian cancer. Our study includes eight full-
thickness fallopian tube samples, but no ovarian surface epithelial samples. The
presence of other cell types in the full-thickness samples may introduce DNA
methylation profiles that differ from the normal epithelial comparator tissue. A
59
technical concern is that all of the fallopian samples were run in a single analysis
batch (batch 9), and the clinical parameters of the tumor samples within this batch
are not representative of those in the entire study. We focused on identifying genes
with large DNA methylation and gene expression differences between fallopian tube
and tumors, to address concerns regarding the small sample size of the comparison
control group, the confounding of biology and batch-associated measurement
biases, which cannot be fully removed using conventional normalization
approaches, and variations in stromal contamination among the tumors.
Third, epigenetic silencing of different genes is likely to occur at varying
frequencies in the tumor data set. To capture genes with a low frequency of
epigenetic silencing, we focused on tumors with high levels of DNA methylation at
that locus, by comparing the 90th percentile of tumors to the mean of the fallopian
tube samples.
We describe the strategy for identifying epigenetically silenced genes in
detail in the methods section. In brief, we apply four separate filtering criteria: 1) low
mean DNA methylation in fallopian tube samples, 2) large difference in DNA
methylation between the 90
th
percentile tumor and mean fallopian tube methylation,
3) large difference in mean gene expression between the fallopian tubes, and the
10% of tumor samples with the highest DNA methylation for that gene, and 4) strong
inverse correlation between DNA methylation and gene expression. For each filter,
we established a relaxed threshold and a stringent threshold. To set minimal criteria
for each filter, while capturing genes with different silencing patterns and
frequencies, we required candidate epigenetically silenced genes to pass all four
relaxed thresholds, and at least three out of four more stringent thresholds.
60
The list of genes resulting from our analysis should be considered a
preliminary list of epigenetic silencing candidates. This list will have likely missed
some important, functionally relevant silenced genes, while including others
inappropriately. These candidate genes for which we have observed correlative
Figure 2-1: BRCA1 gene expression versus promoter methylation. The color and size
of the dots represent tissue type (red/large – fallopian tube samples, n=8; other
colors/small – ovarian tumors, n=489. Specifically, blue dots represent tumors with
BRCA1 epigenetically silencing; green dots represent tumors with BRCA1 germline
mutation; purple dots represent tumors with BRCA1 somatic mutation. Unsequenced
tumors were shown with hollow dots). Plotted in the y-axis is the relative mRNA
expression level of BRCA1 as log ratios reported in the median-integrated expression
data set, and in the x-axis is the DNA methylation beta value.
61
evidence for epigenetic silencing, will require experimental validation by promoter
methylation cassette analysis or DNA methyltransferase inhibitor treatment of cell
lines. A complete list of the 168 genes with evidence for epigenetic silencing is
provided in Appendix A.
BRCA1 is one of the 168 genes identified with this method. A scatterplot of
DNA methylation versus gene expression is shown in Figure 2-1. BRCA1 silencing
via promoter hypermethylation has been reported previously in breast and ovarian
cancer (Esteller et al., 2000b), and recent studies have reported BRCA1
hypermethylation in varying percentages of ovarian cancer patients, but mostly
within 10-20% (Baldwin et al., 2000; Esteller et al., 2000a; Press et al., 2008). With
the procedure described in the methods below, we identified 56 out of 489 samples
(11.5%) with BRCA1 inactivation via promoter hypermethylation in the current high-
grade, high stage serous ovarian cancer cohort (indicated by blue dots in Figure 2-
1). We validated the BRCA1 promoter hypermethylation with the MethyLight
technology (Eads et al., 2000; Eads et al., 1999). MethyLight is a real-time PCR
based method for DNA methylation quantification. The MethyLight PMR value (See
Methods) for BRCA1 showed strong correlation (Pearson Correlation Coeffient =
0.78-0.90) with the beta values measured by the four BRCA1 probes (Figure 2-2). A
receiver operating characteristic (ROC) curve (Figure 2-3) showed that the PMR
values correlate very well with the BRCA1 epigenetic silencing calls, with an AUC of
0.99. Using PMR>10 as the cutoff for promoter hypermethylation, as described by
Weisenberger et al (Weisenberger et al., 2006), MethyLight confirms the promoter
hypermethylation in 55 of the 56 samples (98.2%) previously identified on the
Infinium platform, and the absence of such methylation in 436 of the 441 samples
62
(98.9%), including 433 tumors and 8 normal fallopian tube samples previously
identified to be negative of BRCA1 epigenetic silencing. Overall, the two methods
Figure 2-2: BRCA1 epigenetic silencing validation. The scatterplots show pairwise
comparison of the Infinium beta values and MethyLight PMR values for the 489 ovarian
serous adenocarcinomas. Left four columns (upper four rows) are the four BRCA1 probes
used for making the epigenetical silencing calls. The fifth column(row) shows the PMR
values given by MethyLight. The lower left panels show the pairwise comparison for each
of the five measurements. Each dot represents a sample. The red line indicates a Loess
regression fit (alpha=1.2). The numbers at the upper right panels show the Pearson's
Correlation Coefficient of the two measurements at each intersection. The Infinium probes
and MethyLight probe are arranged by genomic location. All five are located in the same
CpG Island that flanks BRCA1 transcription start site by the Takai Jones definition.
63
showed concordance on 491 of the 497 samples (98.8%), and confirmed the
observed BRCA1 epigenetic silencing.
Notably, BRCA1 epigenetic silencing is mutually exclusive with all BRCA1/2
mutations (P=4.4*10
-4
, Fisher’s exact test). A previous population based study
showed that BRCA1 epigenetic silencing was only seen in ovarian cancer patients
Figure 2-3: ROC curve of BRCA1 methylation validation by MethyLight. The ability of
MethyLight measurement of BRCA1 methylation to discriminate BRCA1 epigenetic
silenced cases from non-silenced cases, as determined by the Illumina Infinium DNA
methylation and gene expression measurements (see methods) is depicted as an ROC
curve. ( AUC=0.99).
64
without a family history associated with a breast/ovarian cancer syndrome (Baldwin
et al., 2000), suggesting that BRCA1 promoter hypermethylation is unlikely to be
inherited, but rather an acquired somatic change that leads to BRCA1 inactivation in
sporadic ovarian cancers. BRCA1 hypermethylated cases are considerably younger
and occur more frequently than the BRCA1 somatic mutation cases. This suggests
Figure 2-4: RAB25 gene expression versus promoter methylation. The color and
size of the dots represent tissue type (red/large – fallopian tube samples, n=8;
black/small – ovarian tumors, n=489).
65
that epigenetic silencing of BRCA1 might be a more efficient somatic mechanism of
inactivation for this gene than mutation. BRCA1/2 mutation cases, especially
BRCA2 mutation cases, are associated with better survival, while BRCA1
epigenetically silenced cases have worse survival.
RAB25 (Figure 2-4) is ranked highest among the 168 genes, based on the
DNA methylation-expression correlation. Previously, RAB25 was reported to have a
Figure 2-5: AMT gene expression versus promoter methylation. The color and size
of the dots represent tissue type (red/large – fallopian tube samples, n=8; black/small –
ovarian tumors, n=489).
66
>1.3-fold copy number increase in about half of advanced serous epithelial ovarian
cancers and marked mRNA up-regulation in most of ovarian cancers, compared to
normal ovarian epithelium, and the copy number and expression levels of RAB25
were associated with disease-free survival or overall survival in ovarian and breast
Figure 2-6: SPARCL1 gene expression versus promoter methylation. The color and
size of the dots represent tissue type (red/large – fallopian tube samples, n=8;
black/small – ovarian tumors, n=489).
67
cancers (Cheng et al., 2004). RNA interference targeting RAB25 has been showed
to slow down cell proliferation and inhibit tumor growth in in vivo and in vitro ovarian
cancer models (Fan et al., 2006). Other papers also highlight the role of RAB25 in
cancer development. Somewhat contrary to these reports on ovarian and breast
cancers, a recent paper (Nam et al., 2010) indicated that loss of RAB25 promotes
intestinal neoplasia, and is associated with human colorectal adenocarcinomas. Our
study did not observe significant amplification of this region. On the contrary, our
results indicate that RAB25 down-regulation actually occurs in a subset of ovarian
Figure 2-7: CCL21 gene expression versus promoter methylation. The color and size
of the dots represent tissue type (red/large – fallopian tube samples, n=8; black/small –
ovarian tumors, n=489).
68
tumors (Figure 2-4). This result, in line with the reported RAB25 loss in intestinal
neoplasia, suggests that loss of RAB25 might play a role in ovarian tumorigenesis.
Among the 168 genes, AMT (Figure 2-5), SPARCL1 (Figure 2-6) and CCL21
(Figure 2-7), are also noteworthy because they show promoter hypermethylation in
the vast majority of tumors. SPARCL1, a member of the SPARC family and anti-
adhesive extracellular matrix protein, was originally shown to be down-regulated in
many epithelium-derived cancers (Bendik et al., 1998; Nelson et al., 1998). A gene
closely related to SPARCL1, SPARC, has been shown to have tumor-suppressor
activity in human ovarian epithelial cells (Mok et al., 1996), and one driver mutation
of SPARC was observed in our study. Loss of SPARCL1 expression has been
shown to be associated with increased proliferation and cell cycle progression
(Claeskens et al., 2000), highlighting its role in tumorigenesis. CCL21 has been
shown to be a chemoattractant for T cells and dendritic cells (Cyster, 1999). Anti-
tumor properties of this gene have been attributed to its role as a chemo-attractant
(Sharma et al., 2003) and as an angiostatic modulator (Vicari et al., 2000).
Discovery of DNA methylation subgroups
Using the resampling-based consensus clustering method as previously
described (Monti et al., 2003), we identified four DNA methylation clusters (Figure 2-
8). However, there is no clear evidence for the existence of a well-defined CpG
Island Methylator Phenotype (CIMP), as has been identified for colorectal carcinoma
(Hinoue et al., 2012; Toyota et al., 1999) and glioblastoma (Noushmehr et al.,
2010), characterized by concerted hypermethylation at CpG islands. There is a
moderate, but statistically significant overlap between DNA methylation clusters and
gene expression subtypes (p<2.2*10
-16
,
€
χ
2
test. Table 2, Adjusted Rand Index =
69
0.07).
Patients belonging to the four clusters differ significantly in age at diagnosis
(One-way ANOVA, p=5*10
-7
) (Figure 2-9B). The mean ages of the patients in the
four clusters are 59.1, 65.8, 57.1, and 62.1 for MC1, MC2, MC3, and MC4,
Figure 2-8: DNA methylation subtypes. Consensus clustering was performed on 489
serous ovarian tumor samples with 858 Infinium probes, selected as described in
Supplemental Methods. DNA cluster membership was determined by 1,000 resampling
iterations of consensus clustering using the K-means algorithm. Hierarchical clustering of
the 192 most discriminant probes is shown in the heatmap, with eight fallopian tube
samples shown on the left. DNA methylation levels (beta value) are shown with a color
spectrum as indicated in the color key panel, with blue indicating no methylation (beta
value=0), to red, indicating full methylation (beta value=1). White indicates missing value.
DNA methylation cluster memberships of the tumors are indicated by the color bar: blue,
Cluster MC1 (n=131); green, Cluster MC2 (n=64); red, Cluster MC3 (n=156), purple,
Cluster MC4 (n=138). Other color bars indicate various molecular features as indicated in
the color key. There is no association between the DNA methylation clusters and
analytical batch (bottom bar, p=0.85,
€
χ
2
test)
70
respectively. Tukey HSD test revealed that the real differences lie between clusters
MC1 and MC2 (adjusted p=0.0005), between MC2 and MC3 (adjusted p=0.000002),
and between MC3 and MC4 (adjusted p=0.0009).
Patients in the four DNA methylation clusters also differ significantly in overall
survival with data censored at five years (Median survival time: Cluster MC1 – 48.9
months, Cluster MC2 – 35.8 months; Cluster MC3 – 40.9 months; Cluster MC4 –
43.6 months; Logrank test, p=0.04.) (Figure 2-9A). After adjusting for age using the
Cox regression model, Cluster MC1 has the best survival and Cluster MC3 has a
significantly worse survival compared to Cluster MC1 (Hazard Ratio = 1.43, p=0.04).
Cluster MC2 has marginally significantly worse age-adjusted survival (Hazard Ratio
= 1.42, p=0.09) than MC1.
Figure 2-9: DNA methylation subtypes – clinical implications. Clinical relevance of
the DNA methylation clusters. A. Kaplan-Meier curves showing the differential survival of
the four DNA methylation clusters with five-year censored survival data. Samples are
colored according to their cluster membership as described in Supplemental Figure 2.
The four clusters differ in overall survival (Median survival time: Median survival time:
Cluster MC1 – 48.9 months, Cluster MC2 – 35.8 months; Cluster MC3 – 40.9 months;
Cluster MC4 – 43.6 months; Logrank test, p=0.04.) B. The distributions of age at
diagnosis for patients in the three DNA methylation clusters are shown in the box-plots,
and patients in the three DNA methylation clusters differ in age at diagnosis (One-way
ANOVA, p=5*10-7). Tukey HSD test revealed that patients in cluster MC2 are an average
of 6.7 years older than the patients in cluster MC1 (95% CI: 2.33-11.15; mean age: 65.8
v.s. 59.1 years; adjusted p=0.0005) and 8.7 years (95% CI: 4.4 -13.0 years; mean age:
65.8 v.s. 57.1 years; adjusted p=0.000002) older than cluster MC3 (and 57.1 years,
adjusted p=0.0005 and 0.000002 respectively), and Cluster MC4 patients are 5.0 years
older than patients belonging to cluster MC3 (95% CI: 1.6 – 8.4 years; mean age: 62.1
v.s. 57.1 years, adjusted p=0.0009).
71
Table 2-1: Distribution of BRCA inactivation events across promoter methylation
clusters.
The four DNA methylation clusters differ significantly in their frequencies of
BRCA inactivation events (p=4.9*10
-6,
Fisher’s exact test), which include BRCA1/2
mutation and BRCA1 epigenetic silencing. Altogether, Cluster MC1 and MC3 have
the highest frequencies of BRCA inactivation (46.6% and 44.5%, respectively), while
Cluster MC2 has the lowest such frequency (13.2%, Table 1). This trend holds true
DNA Methylation
Cluster
MC1 MC2 MC3 MC4 TOTAL
P
value
Total
Sample
set
All Samples 131 64 156 138 489 - -
Sequenced Samples 80 38 115 83 316 - -
Samples With Known
BRCA Status* 88 38 128 84 338 - -
All BRCA Inactivtion
Events 41 5 57 17 120
4.9E-
06 338
All BRCA Inactivtion
Events (%) 46.6 13.2 44.5 20.2 35.5 - 338
BRCA1 Epigenetic
Silencing 19 1 30 6 56
1.0E-
05 489
BRCA Mutation** 22 4 27 11 64 0.04 338
All Germline
Mutations 18 2 20 6 46 0.03 338
BRCA1 Germline
Mutation 11 1 11 4 27 0.19 338
BRCA2 Germline
Mutation 7 1 10 2 20 0.17 338
All Somatic
Mutations 4 2 8 5 19 0.97 338
BRCA1 Somatic
Mutation 1 2 7 0 10 0.05 338
BRCA2 Somatic
Mutation 3 0 1 5 9 0.09 338
* 316 sequenced cases + 22 samples with epigenetic silencing but no sequencing data;
due to the known mutual exclusivity observed, we assume no mutation events in those
22 samples.
** Two cases in Cluster MC3 have both more than one mutations: TCGA-20-1684 has
BRCA2 germline mutations and BRCA1 somatic mutation; TCGA-13-1501 has BRCA2
germline mutations and BRCA1 germline mutation; Otherwise all inactivation events are
mutually exclusive.
72
when we stratify the BRCA inactivation events to epigenetic silencing (p=1.0*10
-5
,
Fisher’s exact test) and BRCA1/2 mutation events (p=0.04, Fisher’s exact test).
Although the four DNA methylation clusters differ in their DNA methylation
profiles and their biology, the overall average silhouette width (Rousseeuw, 1987) is
poor (0.02), indicating weakly defined clusters with substantial within-group
heterogeneity. Indeed, other clustering methods yield varying subgroupings of the
tumors based on their DNA methylation profiles. Therefore, the DNA methylation
cluster memberships reported here should be considered preliminary, with
alternative groupings possible. Nevertheless, the significant overlap with expression
clusters, and other biological differences between the DNA methylation clusters
suggest some validity to these subdivisions.
Table 2-2: Overlap between gene expression and DNA methylation subtypes.
Differentiated Immunoreactive Mesenchymal Proliferative Total
DNA
Methylation
Subtype 1
(MC1)
55 31 24 21 131
DNA
Methylation
Subtype 2
(MC2)
3 2 10 49 64
DNA
Methylation
Subtype 3
(MC3)
41 32 60 23 156
DNA
Methylation
Subtype 4
(MC4)
36 42 15 45 138
Total 135 107 109 138 489
73
Batch Effects In TCGA DNA Methylation Data
Prevalent batch effects were observed in the DNA methylation dataset
(affecting 78.8% of the features, F test), consisting of 489 tumor samples, as also
discussed by Leek et al (Leek et al., 2010). Principal component analysis (PCA)
shows that the second PC is significantly associated with batch (adjusted R-square
= 0.59).
A natural solution would be to employ normalization tools to remove the
technical variation associated with batch. However, DNA methylation data
expressed as beta values do not fit the assumptions for existing normalization
methods developed for gene expression, such as quantile normalization and loess
normalization (Laird, 2010). Most of these methods assume that most genes are not
expressed and/or that most genes are not differentially expressed. However, neither
assumptions hold for DNA methylation data, where global hypomethylation
observed in many cancers may affect a majority of probes. Moreover, samples can
differ substantially in their total methylcytosine content. Quantile normalization would
erase such differences. Meanwhile, for beta-distributed data like DNA methylation
beta values, the variance is associated with the mean (heteroscedasticity).
Therefore, we cannot apply linear model-based methods without transforming the
data properly (logit or probit). This heteroscedasticity also hinders the
straightforward application of empirical Bayes normalization methods like ComBat
(Johnson et al., 2007) since the variances of different probes cannot be modeled
assuming they are the same. In other words, there is a lack of a reasonable prior.
74
In addition to the problems discussed above, there are also substantial
biological differences across batches. For example, five-year survival rates (Logrank
test, p=0.04, for differences across batches) vary from 0% (Batch 9) to 58% (Batch
11). The known or unknown confounding biological differences between batches
may result in removing biology in the normalization process. While introducing a
design matrix to adjust for the known biological factors is feasible, the unknown
confounding biological differences present a more serious problem. While well-
established methods to remove unknown factors such as surrogate variable
analysis (Leek and Storey, 2007) are outstanding methods in dealing with unknown
factors, these will actually be undesirably harsh for unsupervised analyses aimed
at finding molecular subtypes. However, we advise the use of this method in
supervised analyses with TCGA data. This confounding biology reemphasizes the
importance of randomized experimental design in high-throughput studies, as has
been emphasized by Verdugo et al and Leek et al (Leek and Storey, 2007; Verdugo
et al., 2009).
Although most of the probes are susceptible to batch effects, the size of the
batch variation is small. The mean absolute beta values difference across pairwise
comparison of different batches is 0.030 (median difference 0.026) for the probes
that varied significantly across batches. Given the complications discussed above,
we chose not to normalize the data in a way that might remove biological
differences or introduce artifacts, but rather focus our analyses on probes with large
biological variation and limited technological variation. For the clustering analysis,
we removed probes with relatively large batch variation using technical replicates
(weighted average of deviation from equality of >0.05, as described in the method
75
section). By combining this filtering approach with a selection of the most variant
probes, we were able to obtain a dataset in which the large biological variation
predominates over the weak technical variation. None of the top ten principal
components (PCs) were associated with batch in the reduced data set used for
clustering. Cohen’s f
2
(Cohen, 1992) for the 858 probes range from 0.004 to 0.183,
with a mean of 0.029 (1
st
quartile: 0.018, 3
rd
quartile: 0.035). 857 probes out of 858
have an effect size of smaller than medium (rule of thumb 0.15; equivalent to a
multiple regression R of 0.13). On average, only about 2.8% of the variance for each
probe is explained by batch in the dataset used for clustering. We can also see from
the bottom side bar in Figure 2-9 that analytical batch is not driving the clustering.
For the epigenetically silenced genes, we used large effect size as the threshold,
rather than statistical significance. Here, we also observed that batch effects do not
have a deterministic role in the epigenetically silenced genes identified. We would
advise those who use the TCGA DNA methylation data to be aware of the technical
variation in the data, and to consider developing normalization methods appropriate
for beta-distributed DNA methylation data.
76
Methods
DNA Methylation Assays
We performed the Illumina Infinium DNA methylation assay on 519 TCGA
ovarian samples and eight fallopian tube samples from Batches 9,11-15 and 17-19,
21-22, and batch 24. The Illumina Infinium HumanMethylation27 arrays interrogate
27,578 CpG sites located in proximity to the transcription start sites of 14,475
consensus coding sequencing in the NCBI Database (Genome Build 36). Bisulfite
conversion was performed on 1 µg genomic DNA from each patient using the Zymo
EZ96 kit (Zymo Research, Orange, CA) as recommended by the manufacturer. We
evaluated DNA quantity and completeness of bisulfite-conversion using MethyLight
control quality control (QC) reactions as previously described (Campan et al., 2009).
All TCGA samples passed these QC tests and entered the Infinium DNA
methylation assay pipeline.
Bisulfite-converted DNA was whole genome amplified and enzymatically
fragmented. The bisulfite-converted, WGA-DNA samples were then purified and
hybridized to the BeadChip arrays, in which bisulfite-converted DNA molecules
anneal to locus-specific DNA oligomers that are bound to individual bead types.
Each CpG locus can hybridize to methylated (CpG) or unmethylated (TpG) oligo
bead types. DNA methylation-specific primer annealing is followed by single-base
extension using labeled nucleotides. Both unmethylated and methylated bead types
for a specific CpG locus incorporate the same labeled nucleotide, as determined by
the base immediately preceding the cytosine being interrogated by the assay, and
subsequently will be detected in a single channel. Each beadchip, containing 12
77
subarrays, was then fluorescently stained after extension, scanned, and the
intensities of the methylated (M) and unmethylated (U) bead types for each CpG
locus across all samples are measured. Mean non-background corrected M and U
signal intensities for each locus were extracted from Illumina BeadStudio (or
GenomeStudio) software. The beta value DNA methylation scores for each sample
and locus were calculated as (M/(M+U)).
Detection p-values were calculated by comparing the set of analytical probe
replicates for each locus to the set of 16 negative control probes. The negative
controls are modeled by normal distribution. The detection p-value for the probe with
intensity I
probe
, is calculated as: 1-Z( (|I
probe
-µ
neg
|)/σ
neg
). In this formula, µ
neg
and σ
neg
are the average and the standard deviation of the signals from the negative controls,
and Z is the one-sided tail probability of the standard normal distribution. For each
probe, the detection p-values are calculated separately for methylated and
unmethylated probe signal intensities, and the smaller detection p-value was taken
as the final detection p-value for the probe. Data points with detection p-values >
0.05 were deemed not significantly different from background, and were masked as
"NA".
TCGA Data Packages
The data levels and the files contained in each data level package are
described below and are present on the TCGA Data Portal website
(http://tcga.cancer.gov/dataportal).
LEVEL 1: Level 1 data contain the non-background corrected signal
intensities of the methylated (M) and unmethylated (U) probes and the mean
negative control cy5 (red) and cy3 (green) signal intensities. A detection p-value for
78
each data point, the number of replicate beads for methylated and unmethylated
bead types as well as the standard error of methylated and unmethylated signal
intensities are also provided for each sample and probe. Similar values are also
provided for the negative control probes. It is important to note that the identity of
the dye is representative of the nucleotide adjacent to the CpG dinucleotide. The
methylation discrimination is derived from separate measurements from the two
different types of beads present for each locus. For some loci, both measurements
will be cy3, and for others both will be cy5. To resolve ambiguities regarding this
subtlety of the Infinium DNA Methylation assay, we have labeled the cy3 and cy5
values deposited as level 1 data to the TCGA Data Coordination Center (DCC) as
“Methylated Signal Intensity” and “Unmethylated Signal Intensity”. The information
of which dye is used for each locus is supplied in the manifest deposited with the
DCC.
LEVEL 2: Level 2 data files contain the beta value calculations for each
probe and sample. Data points with detection p-values > 0.05 were deemed not
significantly different from background, and were masked as NA.
LEVEL 3: Level 3 data contain beta value calculations, gene IDs and
genomic coordinates for each probe on the array. In addition, data for probes that
contain known single nucleotide polymorphisms (SNPs) after comparison to the
dbSNP database (Build 128) and data for probes that contain repetitive element
DNA sequences in more than 10 bp of each 50 bp probe sequence are masked with
an “NA” descriptor.
The data packages used for the following analyses are listed below. Please
note that with continuing updates of genomic databases, data archive revisions
79
become available at the TCGA data portal. The following data archives were used
for the analyses described in this manuscript:
Batches 9 and 11: jhu-usc.edu_OV.HumanMethylation27.Level_3.1.3.0
(Batches 9 and 11 included all fallopian tube samples)
Batch 12: jhu-usc.edu_OV.HumanMethylation27.Level_3.2.3.0
Batch 13: jhu-usc.edu_OV.HumanMethylation27.Level_3.3.3.0
Batch 14: jhu-usc.edu_OV.HumanMethylation27.Level_3.4.2.0
Batch 15: jhu-usc.edu_OV.HumanMethylation27.Level_3.5.1.0
Batch 17: jhu-usc.edu_OV.HumanMethylation27.Level_3.6.1.0
Batch 18: jhu-usc.edu_OV.HumanMethylation27.Level_3.7.1.0
Batch 19: jhu-usc.edu_OV.HumanMethylation27.Level_3.8.1.0
Batch 21: jhu-usc.edu_OV.HumanMethylation27.Level_2.9.0.0
Batch 22: jhu-usc.edu_OV.HumanMethylation27.Level_2.10.0.0
Batch 24: jhu-usc.edu_OV.HumanMethylation27.Level_2.11.0.0
During the course of data production, the platform manifest was updated to
reflect current HUGO gene symbols, and to mask probes containing recently
identified SNPs or repeats. This manifest update started with Batch 21. To ensure
consistent gene symbol usage across all batches within this study, we used Level 2
data for batches 21, 22, and 24, and generated Level 3 data for these batches using
identical procedures as for the earlier batches (masking the same probes, and
reconciling gene symbols).
Cancer-associated epigenetic silencing
We used Level 3 DNA methylation data on 23,679 DNA methylation probes,
and the median-integrated gene expression data set on 18,868 genes. The median
80
based integrated expression data set was assembled using row-centered Level 3
data generated on the LBL-HuEx, UNC-Agilent and Broad-U133A platforms. This
data set included every gene and every samples that has been profiled on one of
these platform. If a gene was only assayed on one platform (n=1,116), this
measurement was used. If the gene was assayed on two platforms (n=5,890), the
average of the two measurements was used; if the gene was assayed on all
platforms (gene on all three platforms n=11,864) the median measurement was
used. This data set contains 541 samples (including 8 fallopian tube samples) and
18,868 genes. A set of 21,273 Infinium probes that interrogate 12,233 genes have
matched gene expression data. We determined the Spearman correlation between
DNA methylation and gene expression for 497 samples (including 489 tumor
samples and eight fallopian tube samples. We used the non-parametric Spearman
method, as bivariate normality could not be assumed (DNA methylation data are not
normally distributed). Spearman's rank correlation coefficient (ρ) on the gene
expression and DNA methylation was computed for each probe, along with a p
value testing against the null hypothesis that ρ truly equals zero. The Benjamini-
Hochberg procedure was used to control the false discovery rate. Given the small
sample size for the fallopian tubes, we excluded 49 probes that failed on any of the
fallopian tube samples, with 23,630 DNA methylation probes remaining, of which
21,229 covering 12,206 genes, could be matched to expression.
We describe below the strategy used to identify candidate epigenetically
silenced genes. We apply four separate filtering criteria. The thresholds for each
filter were selected based on inspection of scatterplots of DNA methylation versus
81
gene expression for the genes passing the relevant filter criterion in the current data
set. The four filters are:
1) The sample mean beta value of eight normal fallopian tubes with a
relaxed threshold of 0.5 and a stringent threshold of 0.4.
2) The difference in DNA methylation between the 90
th
percentile tumor and
mean fallopian tube methylation, with a relaxed threshold of 0.1 and a stringent
threshold of 0.3.
3) The fold difference in mean gene expression between the fallopian tubes,
and the 10% of tumor samples with the highest DNA methylation for that gene, with
a relaxed threshold of 2-fold and a stringent threshold of 3-fold.
4) Spearman’s correlation coefficient between DNA methylation and gene
expression calculated jointly across 489 tumor and 8 fallopian samples, with a
relaxed ρ threshold of –0.2 and a stringent threshold of –0.3.
We required candidate epigenetically silenced genes to pass all four relaxed
thresholds, and at least three out of four more stringent thresholds. If there were
multiple CpGs for the same gene promoter, the CpG with the highest absolute
Spearman’s Rho was retained for that gene. A complete list of the 168 genes is
shown in Appendix A, ranked by descending absolute Spearman’s Rho.
Definition of BRCA1 epigenetically silenced cases
We analyzed the relationship between DNA methylation and gene
expression for nine different probes located in or near the BRCA1 promoter region,
and found statistically significant inverse correlations for four of the nine probes
(cg19531713, cg19088651, cg08993267, cg04658354). The target CpG sites of
those probes are located in the CpG island that contains the transcription start site
82
of BRCA1. The Spearman ρ of these correlations ranges from –0.28 to –0.37
(Benjamini-Hochberg adjusted P<0.0001). We did not see good inverse correlations
for the other five probes located in other two CpG islands further away from the
transcription start site.
For each of the aforementioned four probes, we used K-means clustering
(assuming K=2) on the two-dimensional space of DNA methylation and expression
data to separate the epigenetically silenced group and non-epigenetically silenced
group of samples. Expression data were scaled to have the same range as DNA
methylation data for this clustering. We then combined the calls from the four
probes. Since data was lacking for some probes in some samples, we relied on the
fraction of the four probes calling a particular sample epigenetically silenced for
BRCA1, rather than on a fixed number of probes. Samples with >50% consensus on
belonging to the epigenetically silenced group across the four probes were classified
as samples with silencing of BRCA1 by promoter hypermethylation.
Validation of BRCA1 epigenetic silencing with MethyLight
The Infinium DNA Methylation data for the BRCA1 promoter was valiadated
for all 489 TCGA ovarian serous adenocarcinoma samples using MethyLight
technology (Campan et al., 2009; Eads et al., 2000; Eads et al., 1999). The BRCA1-
M1 MethyLight assay (HB-045) utilized primer and probe oligomers described
previously (Fiegl et al., 2004). MethyLight data are reported as a ratio between the
value derived from the real-time PCR standard curve plotted as log (quantity) versus
threshold C(t) value for the BRCA1 methylation reaction and likewise for a
methylation-independent control reaction based on interspersed ALU repeats
(Weisenberger et al., 2005; Weisenberger et al., 2006). This calculation was
83
performed for both the sample and an M.SssI-treated genomic DNA sample, which
was used as a methylated reference. We calculated the percent of methylated
reference (PMR) for each sample as: 100 X (BRCA1-M1 / ALU)sample / (BRCA1-
M1 / ALU)M.SssI-Reference.
Unsupervised clustering analysis for DNA methylation subtype discovery
We performed unsupervised clustering analysis on 489 high-grade, high-
stage TCGA samples based on DNA methylation data. We first removed 3,899
probes containing a single nucleotide polymorphism (SNP) within five base pairs of
the target CpG site and those containing repeat element sequences of ≥10 base
pairs.
Most normalization methods developed for gene expression arrays are not
suitable for Infinium DNA methylation data for the following reasons: 1) The Cy3 and
Cy5 dyes are not tied to the methylation status of the probes, 2) The majority of loci
cannot be assumed to be unmethylated, 3) The total signal or methylation levels of
different samples cannot be assumed to be equal, and 4) The measurements of the
methylated and unmethylated probes are not independent. Moreover, we observed
significant differences in biology and clinical parameters between analysis batches
of samples. Therefore, rather than attempt to dissociate biology from technical batch
effects through normalization, we chose to rely on robust probes for the
unsupervised clustering, by eliminating probes that introduce technical noise. We
compared technical replicates of the same sample (TCGA-07-0227) that were
measured across ten different batches, to identify probes that are prone to variation
across batches. The underlying assumption is that technical replicates should yield
identical beta values for each probe. We determined the deviation from this
84
assumption for each probe by calculating the distance of point (x,y) to y=x, in which
x is the beta value of that probe for the same sample in one batch, and y the beta
value for the other batch. Mathematically, this distance (D) can be calculated as
D=(y-x)/√2) for each probe. We then calculated the rank-weighted mean (with
penalty for probe failures) of the distances (D) calculated from 45 pair-wise
comparisons for each of the 23,679 probes. To exclude as much technical noise as
possible, we removed 10,589 probes with a weighed mean D greater than 0.05. Of
the remaining 12,990 probes, we selected the most variant 858 probes for the
clustering analysis. The variant probes were selected as the union of top 5% of
probes with the largest standard deviation and top 5% with the largest adjusted
standard deviation (σ’) normalized for the Bernoulli distribution standard deviation
for the associated mean (σ’=σ/(√(µ(1-µ)))), since the maximal standard deviation of
a beta distribution is influenced by its mean, and equals the standard deviation of a
Bernoulli distribution.
Ovarian DNA methylation subtypes were discovered using consensus
clustering (GenePattern, v 3.2.3.) (Reich et al., 2006). The optimal number of
clusters was determined with 1,000 resampling iterations (seed value: 12,345) using
K-means clustering algorithm for K=2,3,4,…,10, with Euclidean distance as the
distance measure.
We developed signatures for the clusters by selecting the top 50 probes for
each cluster compared to the other clusters. The union of the probes (192 unique
probes) was then clustered on the 489 samples to generate Figure 2-8. The R
software (version 2.11.1) (http://www.r-project.org) was used for all data analyses.
85
Batch Effect Investigation
A fast singular value decomposition (SVD) done with the corpcor R package
was used to extract the principal components (PCs). Each of the top ten PCs was
tested for association with batch using linear regression. A univariate F test was
used to test for the association of each probe and analytical batch. The Benjamini-
Hochberg method was used to adjust for multiple comparisons and control the false-
discovery rate. Cohen’s f
2
was used to assess the effect size of the batch effect.
Other statistical analyses
Pearson’s χ
2
test was used to assess the unequal distributions of categorical
outcomes (e.g., differences in the frequencies of BRCA1 inactivation events across
DNA methylation clusters), with DNA methylation clusters. For covariates with fewer
than five observations in any cell in the R*C contingency table, Fisher’s exact test
was used instead. A Logrank test was used to test against the null hypothesis that
there was no difference between the Kaplan-Meier survival curves. Proportional
hazards regression (Cox Regression) was used for parametric analysis to estimate
hazard ratios associated with unit changes in any continuous variable, or the
comparison of survival after adjusting for other variables, or test for an interaction
term between two variables. The differences were considered significant if the two-
sided p values are <0.05. All statistical tests were performed in R (http://www.r-
project.org).
86
Chapter Two Bibliography
American Cancer Society (2013). Cancer Facts & Figures 2013. Atlanta: American
Cancer Society;.
Baldwin, R.L., Nemeth, E., Tran, H., Shvartsman, H., Cass, I., Narod, S., and
Karlan, B.Y. (2000). BRCA1 promoter region hypermethylation in ovarian
carcinoma: a population-based study. Cancer Res 60, 5329-5333.
Bendik, I., Schraml, P., and Ludwig, C.U. (1998). Characterization of MAST9/Hevin,
a SPARC-like protein, that is down-regulated in non-small cell lung cancer. Cancer
Res 58, 626-629.
Campan, M., Weisenberger, D.J., Trinh, B., and Laird, P.W. (2009). MethyLight.
Methods in molecular biology 507, 325-337.
Cheng, K.W., Lahad, J.P., Kuo, W.L., Lapuk, A., Yamada, K., Auersperg, N., Liu, J.,
Smith-McCune, K., Lu, K.H., Fishman, D., et al. (2004). The RAB25 small GTPase
determines aggressiveness of ovarian and breast cancers. Nat Med 10, 1251-1256.
Claeskens, A., Ongenae, N., Neefs, J.M., Cheyns, P., Kaijen, P., Cools, M., and
Kutoh, E. (2000). Hevin is down-regulated in many cancers and is a negative
regulator of cell growth and proliferation. Br J Cancer 82, 1123-1130.
Cohen, J. (1992). A power primer. Psychological bulletin 112, 155-159.
Cyster, J.G. (1999). Chemokines and the homing of dendritic cells to the T cell
areas of lymphoid organs. J Exp Med 189, 447-450.
Eads, C.A., Danenberg, K.D., Kawakami, K., Saltz, L.B., Blake, C., Shibata, D.,
Danenberg, P.V., and Laird, P.W. (2000). MethyLight: a high-throughput assay to
measure DNA methylation. Nucleic acids research 28, E32.
Eads, C.A., Danenberg, K.D., Kawakami, K., Saltz, L.B., Danenberg, P.V., and
Laird, P.W. (1999). CpG island hypermethylation in human colorectal tumors is not
associated with DNA methyltransferase overexpression. Cancer Research 59,
2302-2306.
87
Esteller, M., Silva, J.M., Dominguez, G., Bonilla, F., Matias-Guiu, X., Lerma, E.,
Bussaglia, E., Prat, J., Harkes, I.C., Repasky, E.A., et al. (2000a). Promoter
hypermethylation and BRCA1 inactivation in sporadic breast and ovarian tumors. J
Natl Cancer Inst 92, 564-569.
Esteller, M., Silva, J.M., Dominguez, G., Bonilla, F., Matias-Guiu, X., Lerma, E.,
Bussaglia, E., Prat, J., Harkes, I.C., Repasky, E.A., et al. (2000b). Promoter
hypermethylation and BRCA1 inactivation in sporadic breast and ovarian tumors. J
Natl Cancer Inst 92, 564-569.
Fan, Y., Xin, X.Y., Chen, B.L., and Ma, X. (2006). Knockdown of RAB25 expression
by RNAi inhibits growth of human epithelial ovarian cancer cells in vitro and in vivo.
Pathology 38, 561-567.
Fiegl, H., Gattringer, C., Widschwendter, A., Schneitter, A., Ramoni, A., Sarlay, D.,
Gaugg, I., Goebel, G., Muller, H.M., Mueller-Holzner, E., et al. (2004). Methylated
DNA collected by tampons--a new tool to detect endometrial cancer. Cancer
Epidemiol Biomarkers Prev 13, 882-888.
Hinoue, T., Weisenberger, D.J., Lange, C.P., Shen, H., Byun, H.M., Van Den Berg,
D., Malik, S., Pan, F., Noushmehr, H., van Dijk, C.M., et al. (2012). Genome-scale
analysis of aberrant DNA methylation in colorectal cancer. Genome research 22,
271-282.
Johnson, W.E., Li, C., and Rabinovic, A. (2007). Adjusting batch effects in
microarray expression data using empirical Bayes methods. Biostatistics 8, 118-
127.
Jones, P.A., and Baylin, S.B. (2002). The fundamental role of epigenetic events in
cancer. Nat Rev Genet 3, 415-428.
Jones, P.A., and Laird, P.W. (1999). Cancer epigenetics comes of age. Nature
Genetics 21, 163-167.
Kurman, R.J., and Shih Ie, M. (2010). The origin and pathogenesis of epithelial
ovarian cancer: a proposed unifying theory. The American journal of surgical
pathology 34, 433-443.
88
Laird, P.W. (2010). Principles and challenges of genome-wide DNA methylation
analysis. Nat Rev Genet 11, 191-203.
Leek, J.T., Scharpf, R.B., Bravo, H.C., Simcha, D., Langmead, B., Johnson, W.E.,
Geman, D., Baggerly, K., and Irizarry, R.A. (2010). Tackling the widespread and
critical impact of batch effects in high-throughput data. Nat Rev Genet 11, 733-739.
Leek, J.T., and Storey, J.D. (2007). Capturing heterogeneity in gene expression
studies by surrogate variable analysis. PLoS Genet 3, 1724-1735.
Miller, D.S., Blessing, J.A., Krasner, C.N., Mannel, R.S., Hanjani, P., Pearl, M.L.,
Waggoner, S.E., and Boardman, C.H. (2009). Phase II evaluation of pemetrexed in
the treatment of recurrent or persistent platinum-resistant ovarian or primary
peritoneal carcinoma: a study of the Gynecologic Oncology Group. J Clin Oncol 27,
2686-2691.
Mok, S.C., Chan, W.Y., Wong, K.K., Muto, M.G., and Berkowitz, R.S. (1996).
SPARC, an extracellular matrix protein with tumor-suppressing activity in human
ovarian epithelial cells. Oncogene 12, 1895-1901.
Monti, S., Tamayo, P., Mesirov, J., and Golub, T. (2003). Consensus Clustering: A
Resampling-Based Method for Class Discovery and Visualization of Gene
Expression Microarray Data. 52, 91-118.
Nam, K.T., Lee, H.J., Smith, J.J., Lapierre, L.A., Kamath, V.P., Chen, X., Aronow,
B.J., Yeatman, T.J., Bhartur, S.G., Calhoun, B.C., et al. (2010). Loss of Rab25
promotes the development of intestinal neoplasia in mice and is associated with
human colorectal adenocarcinomas. J Clin Invest 120, 840-849.
Nelson, P.S., Plymate, S.R., Wang, K., True, L.D., Ware, J.L., Gan, L., Liu, A.Y.,
and Hood, L. (1998). Hevin, an antiadhesive extracellular matrix protein, is down-
regulated in metastatic prostate adenocarcinoma. Cancer Res 58, 232-236.
Noushmehr, H., Weisenberger, D.J., Diefes, K., Phillips, H.S., Pujara, K., Berman,
B.P., Pan, F., Pelloski, C.E., Sulman, E.P., Bhat, K.P., et al. (2010). Identification of
a CpG island methylator phenotype that defines a distinct subgroup of glioma.
Cancer Cell 17, 510-522.
89
Press, J.Z., De Luca, A., Boyd, N., Young, S., Troussard, A., Ridge, Y., Kaurah, P.,
Kalloger, S.E., Blood, K.A., Smith, M., et al. (2008). Ovarian carcinomas with
genetic and epigenetic BRCA1 loss have distinct molecular abnormalities. BMC
Cancer 8, 17.
Reich, M., Liefeld, T., Gould, J., Lerner, J., Tamayo, P., and Mesirov, J.P. (2006).
GenePattern 2.0. Nat Genet 38, 500-501.
Rousseeuw, P.J. (1987). Silhouettes: A graphical aid to the interpretation and
validation of cluster analysis. Journal of Computational and Applied Mathematics 20,
53-65.
Seidman, J.D., Horkayne-Szakaly, I., Haiba, M., Boice, C.R., Kurman, R.J., and
Ronnett, B.M. (2004). The histologic type and stage distribution of ovarian
carcinomas of surface epithelial origin. International journal of gynecological
pathology : official journal of the International Society of Gynecological Pathologists
23, 41-44.
Sharma, S., Yang, S.C., Hillinger, S., Zhu, L.X., Huang, M., Batra, R.K., Lin, J.F.,
Burdick, M.D., Strieter, R.M., and Dubinett, S.M. (2003). SLC/CCL21-mediated anti-
tumor responses require IFNgamma, MIG/CXCL9 and IP-10/CXCL10. Mol Cancer
2, 22.
TCGA, C.G.A.R.N. (2011). Integrated genomic analyses of ovarian carcinoma.
Nature 474, 609-615.
Toyota, M., Ahuja, N., Ohe-Toyota, M., Herman, J.G., Baylin, S.B., and Issa, J.P.
(1999). CpG island methylator phenotype in colorectal cancer. Proc Natl Acad Sci U
S A 96, 8681-8686.
Verdugo, R.A., Deschepper, C.F., Munoz, G., Pomp, D., and Churchill, G.A. (2009).
Importance of randomization in microarray experimental designs with Illumina
platforms. Nucleic Acids Res 37, 5610-5618.
Vicari, A.P., Ait-Yahia, S., Chemin, K., Mueller, A., Zlotnik, A., and Caux, C. (2000).
Antitumor effects of the mouse chemokine 6Ckine/SLC through angiostatic and
immunological mechanisms. J Immunol 165, 1992-2000.
90
Weisenberger, D.J., Campan, M., Long, T.I., Kim, M., Woods, C., Fiala, E., Ehrlich,
M., and Laird, P.W. (2005). Analysis of repetitive element DNA methylation by
MethyLight. Nucleic acids research 33, 6823-6836.
Weisenberger, D.J., Siegmund, K.D., Campan, M., Young, J., Long, T.I., Faasse,
M.A., Kang, G.H., Widschwendter, M., Weener, D., Buchanan, D., et al. (2006).
CpG island methylator phenotype underlies sporadic microsatellite instability and is
tightly associated with BRAF mutation in colorectal cancer. Nature Genetics 38,
787-793.
! 91
Chapter Three: Endometrial Cancer
Introduction
Endometrial cancer (uterine corpse cancer) is the most common gynecological
cancer and is the fourth commonest cancer among females following breast, lung and
colon cancers in the US, accounting for 6% of all female cancers. It was estimated that
49,560 new endometrial cancer cases would be diagnosed in the United States and
8,190 would die of the disease in 2013 (American Cancer Society, 2013). The 1- and 5-
year survival rates for endometrial cancer are 92% and 82%, respectively. The 5-year
survival rate is 95% or 67% if the cancer is diagnosed at local or regional stage
respectively. If the disease is diagnosed at advanced stages, however, this number will
drop to 16% (American Cancer Society, 2013). There are two major histological
subtypes – endometrioid and serous, both of which are present for epithelial ovarian
cancer as well. Similar to ovarian cancer, endometrial cancers can also be classified into
Type I or Type II tumors (Lax and Kurman, 1997). Type I tumors occur in younger
women (pre- and peri- menopause), often present low-grade endometrioid histology, and
are estrogen-driven (hormone-receptor positive). This group of tumor is associated with
obesity, most likely via increased estrogen levels, and molecularly with PTEN and KRAS
mutations. Type II tumors, usually of serous or high-grade endometrioid histology, are
more aggressive and more common in older women and are usually negative for
hormone receptors. Different subtypes are believed to have different etiologies, and are
associated with different clinical outcomes.
! 92
Results
Epigenetic Silencing of MLH1
By integrating gene expression and DNA methylation data, we observed frequent
MLH1 promoter hypermethylation in 114 of the 373 samples (30.5%), including 112
endometrioid, 1 mixed and 1 serous tumor(s) respectively (Table 3-1). 37% of the
endometrioid subtype of endometrial cancer has MLH1 epigenetic silencing. This rate is
higher than previously have been reported (Simpkins et al., 1999), even though the
current cohort is biased towards the serous subtype, which usually do not have MSI.
This extensive hypermethylation spans the entire CpG Island associated with MLH1
!
Figure 3-1: MLH1 is methylated across the CpG island associated with its promoter in
about 30% of the tumor samples. Top: transcripts (purple blue), CpG island (green), DNA
methylation level at CpG loci interrogated by HumanMethylation450 probes (light blue, bar
height indicating DNA methylation level from 0 to 1). Bottom left: DNA methylation level for
256 endometrial tumor samples (rows) across the CpG loci interrogated by
HumanMethylation450 probes (column, arranged by genomic location). Blue to red indicate
low (beta=0) to high (beta=1) DNA methylation levels. Samples are clustered based on
MLH1 expression data. Bottom right: MLH1 expression level as measured by RPKM for the
same set of samples, matched to the bottom left panel. Each column is one exon, with the
color of each cell indicating expression level (blue – low expression; red - high expression).
!
! 93
promoter, and abrogates MLH1 expression (Figure 3-1). Promoter methylation of MLH1
is strongly associated with the MSI phenotype (P<10e-16, Table 3-1) featured by a 10-
fold increased mutation rate, with epigenetic silencing of MLH1 accounting for 83% of
the MSI tumors.!
Histology P<0.0001
Endometrioid Mixed Serous
MLH1 Methylated 112 (37%) 1 (8%) 1(2%)
MLH1 Unmethylated 193 (63%) 11 (92%) 52 (98%)
MSI status P<0.0001
Indeterminant MSI-H MSI-L MSS
MLH1 Methylated 0 (0%) 105 (83%) 1 (5%) 8 (4%)
MLH1 Unmethylated 3 (100%) 22 (17%) 19 (95%) 212 (96%)
Table 3-1: MLH1 epigenetic silencing differs by histological subtype, and is strongly
associated with microsatellite instability (MSI). P values are from Fisher’s exact test. MLH1
epigenetic silencing happens in 37% of the endometrioid subtype of endometrial cancers, and
explains 83% of the MSI cases.
DNA Methylation Subtypes of Endometrial Cancer
Unsupervised clustering of DNA methylation data from the 373 endometrial
tumor samples revealed four unique DNA methylation subtypes (MC1-4), typified by a
heavily methylated subtype (MC1) reminiscent of the CpG island methylator phenotype
(CIMP) phenotype described in colon and glioblastoma (Hinoue et al., 2012; Noushmehr
et al., 2010; Toyota et al., 1999), as well as a serous-like cluster (MC3) composed
primarily of serous tumors (Figure 3-2). The CIMP phenotype was associated with the
MSI phenotype, attributable to promoter hypermethylation of MLH1. This association
was similar to the colorectal CIMP but not the glioblastoma CIMP, suggesting a potential
shared mechanism for epithelial CIMP tumors. Indeed, many of the endometrial CIMP
targets, including MLH1, SFRPs and APC, are similar to those observed in colon cancer.
! 94
The endometrial CIMP is not associated with BRAF mutation, the mutation most strongly
associated with colorectal CIMP. MC2 tumors exhibited relatively high cancer-specific
DNA methylation levels, only lower than MC1. DNA hypermethylation observed in MC4
was the lowest among the non-serous-like tumors. MC3, the serous-like cluster, had
minimal DNA methylation changes compared to normal endometrium. MC3 tumors
exhibit extensive copy number changes (Figure 3-3), and correlated well with subtypes
!
Figure 3-2: Unsupervised clustering of the DNA methylation data reveals four
subtypes. A spectrum of blue to red in the heatmap indicates low to high DNA methylation
(0% to 100%). Four DNA methylation subtypes among the 373 tumors (column) are
visualized for 785 CpG loci (row) used for the clustering. Column-side color bars indicate
different features of each sample, the bottom of which shows grouping of the samples as
determined by RPMM (recursively partitioned mixture model for Beta and Gaussian
Mixtures). Within each cluster the samples are seriated by hierarchical clustering. 27 normal
endometrium samples are also plotted for the same loci for comparison. Platform (light blue
– HumanMethylation27; dark blue – HumanMethylation450) and analytical batch for each
sample are plotted on the bottom to make sure that the clustering results are not driven by
technical variations.!
! 95
defined based on mRNA expression and miRNA expression, indicating that they are
indeed a distinct group of tumors.
We also detected decreased methylation level at the X chromosome for most of
the MC3 tumors (Figure 3-4). Three potential scenarios could explain this observation: 1.
Selective loss of the inactive X; 2. Loss of the inactive X (Xi) and replication of the active
X (Xa), as was shown in breast cancer cell lines (Sirchia et al., 2005); 3. Loss of
methylation at the inactive X. In any case, there is an elevated Xa to Xi ratio, indicating
that either the overexpression of a X-linked gene or loss of the inactive X is important for
the tumorigenesis. Preferential loss of the inactive X has been proposed as a
mechanism for loss of XIST expression in ovarian cancer cell lines (Huang et al., 2002),
as XIST is solely expressed from the inactive X chromosome, and the same study
showed that higher XIST level was associated with better response to carbonplatin and
taxol therapy in ovarian cancers.!
!
Figure 3-3. The serous-like tumors (MC3) have extensive copy-number changes.
Copy number profiles for each sample (rows, grouped by DNA methylation subtype) are
visualized in IGV. Blue indicates deletion and red indicates amplification.
! 96
This loss of X-inactivation is also evident in serous ovarian tumors (Figure 3-4).
Suggesting a shared defect between MC3 endometrial tumors and serous ovarian
tumors.
!
!
Figure 3-4: The serous-like tumors also exhibit loss of DNA methylation at X-linked loci.
A spectrum of blue to red in the heatmap indicates low to high DNA methylation (0% to
100%). Four DNA methylation subtypes among the 373 tumors (column) are visualized for
785 CpG loci (top panels) used for the clustering, as well as at all 789 loci (bottom panels)
annotated to the X chromosome. Column-side color bars indicate TP53 mutation status
(black: mutated; gray: wildtype; white - unknown) and DNA methylation cluster membership.
Normal endometrium (left) and serous ovarian cancers (right) are also plotted for the same
loci for comparison.
! 97
Similarity Between MC3 Subtype of Endometrial Cancers and Other Cancers
The MC3 DNA methylation subtype largely overlaps with the tumors with
frequent copy number changes, indicating that those tumors are copy-number instead of
DNA-methylation driven. This feature is very much similar to the serous ovarian tumors.
Indeed, the DNA methylation profile of MC3 is almost identical to that of serous ovarian
tumors (Figure 3-5), which is also similar to basal-like breast tumors (TCGA, 2012c). The
three disease types share features such as minimal DNA methylation change, extensive
copy number change and high frequency of TP53 mutations. They cluster together in a
joint clustering analysis (Figure 3-6) for DNA methylation patterns. However, the MC3
tumors are not entirely identical to their ovarian counterparts. For example, extensive
BRCA1 and BRCA2 epigenetic inactivation through promoter hypermethylation was
reported for both serous OvCa (TCGA, 2011) and basal-like breast (TCGA, 2012c) but
only one out of 81 MC3 samples has BRCA1 methylation. Other known differences exist,
!
Figure 3-5: Similarity between the serous-like endometrial tumors, serous ovarian
tumors, and basal-like breast tumors. 576 serous ovarian tumors (TCGA, 2011) and 146
basal-like breast tumors (TCGA, 2012c) from TCGA are juxtaposed with the endometrial
tumors on the same 785 loci used for clustering for endometrial tumors.
!
! 98
such as the fact that serous endometrial cancer (which makes up the majority of MC3)
does not respond well to chemotherapies while serous ovarian cancers do.
! !
!
Figure 3-6: Cross-tumor comparison. A dendrogram from hierarchical clustering of
different tumors, on the 785 probes used for endometrial cancer clustering. E-MC1/2/3/4:
Endometrial cancer DNA methylation cluster 1-4; B-LumA: the Luminal A subtype of breast
cancer; B-LumB: Luminal B subtype of breast cancer; B-Basal: basal-like subtype of breast
cancer; B-Her2: Her-2 subtype of breast cancer. LUSC: squamous lung cancer;
COADREAD: colorectal cancer; GBM: glioblastoma multiforme. (TCGA, 2008, 2011, 2012a,
b, c)
!
! 99
!
Methods
Sample Collection (The Cancer Genome Atlas UCEC Project)
As part of the Cancer Genome Atlas (TCGA, http://cancergenome.nih.gov/)
Uterine Corpse Endometrial Carcinoma (UCEC) project, tumor samples from 373
patients with newly diagnosed untreated endometrial carcinoma of any of the following
subtypes were collected: endometrioid (n=307), serous (n=53), or mixed (n=13). Clear
cell and other rare subtypes were not included. Local institutional review boards
approved all tissue acquisition. A of May 20, 2012, the median follow-up of the cohort
was 31 months, 21% of the patients have recurred, and 11% have died. Comprehensive
molecular analyses, including mutation, copy number, DNA methylation, gene
expression, miRNA and reverse phase protein array (RPPA) were performed, with DNA
methylation profiled here at USC. The complete datasets are available at https://tcga-
data.nci.nih.gov/tcgafiles/ftp_auth/distro_ftpusers/anonymous/tumor/ucec/. .
Microsatellite instability was identified in 40% of endometrioid tumors and 2% of serous
tumors (TCGA, In Press).
Table 3-2: Distributation of samples by analytical batch and platform. 373 tumors and 27
normal samples were assayed on two platforms. Technical replicates (DNA from the same cell
line) are run together with the samples to control for technical variations.
!
Array-based DNA methylation assay
We used two Illumina Infinium DNA methylation platforms, HumanMethylation27
(HM27) BeadChip and HumanMethylation450 (HM450) BeadChip (Illumina, San Diego,
Batch 49 59 73 75 81 92 94 104 110 118 121 125 137 143 156 sum
Tumor 24 46 47 8 20 47 47 30 22 7 21 6 12 8 28 373 (117+256)
Normal 001 0 0 1 0 4 0 1 0 0 1 8 11 27 (1+26)
Cell Line DNA Replicate 110 1111 1 1 1 1 1 1 1 1 14 (2+12)
Platform HM27k HM450k
1
! 100
CA) to obtain gene promoter and gene body DNA methylation profiles of 373 TCGA
endometrial cancer samples and 27 adjacent non-tumor endometrial tissue samples.
The Infinium HM27 array targets 27,578 CpG sites located in proximity to the
transcription start sites of 14,475 consensus coding sequencing (CCDS) in the NCBI
Database (Genome Build 36). The Infinium HM450 array targets 482,421 CpG sites and
covers 99% of RefSeq genes, with an average of 17 CpG sites per gene region
distributed across the promoter, 5’UTR, first exon, gene body, and 3’UTR. It covers 96%
of CpG islands, with additional coverage in island shores and the regions flanking them.
The assay probe sequences and information on each interrogated CpG site on both
Infinium DNA methylation platforms can be found in the MAGE-TAB ADF (Array Design
Format) file deposited on the TCGA Data Portal.
We performed bisulfite conversion on 1 µg of genomic DNA from each sample
using the EZ-96 DNA Methylation Kit (Zymo Research, Irvine, CA) according to the
manufacturer’s instructions. We assessed the amount of bisulfite converted DNA and
completeness of bisulfite conversion using a panel of MethyLight-based quality control
(QC) reactions as previously described (Campan et al., 2009). All the TCGA samples
passed our QC tests and entered the Infinium DNA methylation assay pipeline.
Bisulfite-converted DNA was whole genome amplified (WGA) and enzymatically
fragmented prior to hybridization to the arrays. BeadArrays were scanned using the
Illumina iScan technology, and the IDAT files (Level 1 data) were used to extract the
intensities (Level 2 data) and calculate the beta value (Level 3 data) for each probe and
sample with the R-based methylumi package.
The level of DNA methylation at each CpG locus is summarized as beta (β) value
calculated as (M/(M+U)), ranging from 0 to 1, which represents the the ratio of the
! 101
methylated probe intensity to the overall intensity at each CpG locus. A p value
comparing the intensity for each probe to the background level was calculated with the
methylumi package at the same time, and data points with a detection P value >0.05
were deemed not significantly different from background measurements, and therefore
were masked as “NA” in the Level 2 and 3 in HM27 and Level 3 in HM450 data
packages, as detailed below.
TCGA data packages
The three data levels are described below and are present on the TCGA Data
Portal website (http://tcga-data.nci.nih.gov/tcga/). Please note that with continuing
updates of genomic databases, data archive revisions become available at the TCGA
Data Portal.
HM27: Level 1 - Level 1 data packages contain the non-background corrected
signal intensities of the M and U probes and the mean negative control cy5 (red) and cy3
(green) signal intensities. A detection P value for each data point, the number of
replicate beads for M and U probes as well as the standard error of M, U, and control
probe signal intensities are also provided. It is important to note that for some CpG
targets, both M and U measurements will be cy3, and for others both will be cy5. To
resolve ambiguities regarding this subtlety of the Infinium DNA Methylation assay, we
have labeled the cy3 and cy5 values deposited to the DCC as “Methylated Signal
Intensity” and “Unmethylated Signal Intensity”. The information of the color channel for
each CpG locus is contained in the MAGE-TAB ADF file deposited in the DCC. Level 2 -
Level 2 data files contain the β-value calculations for each probe and sample. Data
points with detection P values >0.05 were not considered to be significantly different
from background, and were masked as “NA”. Level 3 - Level 3 data contain β-value
! 102
calculations, HUGO gene symbol, chromosome number and genomic coordinate for
each targeted CpG site on the array. In addition, we masked data points with "NA" from
the probes that 1) contain known single nucleotide polymorphisms (SNPs) after
comparison to the dbSNP database (Build 130), 2) contain repetitive sequence elements
that cover the targeted CpG locus in each 50 bp probe sequence, 3) are not uniquely
aligned to the human genome (NCBI build 36.1) at 20 nucleotides at the 3’ terminus of
the probe sequence, 4) span known regions of small insertions and deletions (indels) in
the human genome (dbSNP build 130).
HM450: Level 1 - Level 1 data contain raw IDAT files. IDAT files are the direct
output from the scanning program. Level 2 - Level 2 data contain background corrected
signal intensities of the M and U probes. Level 3 - Level 3 data files contain β-value
calculations and masked data points with "NA" from the probes that are annotated as
having a SNP within 10 base pairs of the interrogated locus (HM27 carryover or recently
discovered). The genomic characteristics for each probe are available for download via
Illumina (www.illumina.com).
Unsupervised clustering analysis of DNA methylation data
The shared probe set between HM27 and HM450 platforms (N=25,978) were
used for this analysis. We removed probes that contained any masked data due to
detection p value, repeats and SNPs and non-uniquely mapped probes (n=22,071
remaining). We observed batch and platform specific effects. To alleviate systematic
platform-specific effects (dye bias, background level, etc) we fit a LOESS regression
model between the two platforms using M values, stratified by the number of CpGs in
the probe (CpG=1,2,3,4,5,6+), and normalized the HM450 data against the HM27 data.
M value is the log 2 ratio of Methylated (M) intensity and Unmethylated (U) intensity and
! 103
better satisifies the linearity assumption. In order to further filter out probes with high
technical variances, we applied a two-way nested ANOVA for platform and batch effects
with batch nested in platform (M value ~ Platform + Platform/Batch) and removed probes
with above-median F value for either platform or batch. We then selected probes with
standard deviation of >1.8 (n=785 probes) based on M values for unsupervised
clustering. Beta values were used for clustering with a mixture model based method,
RPMM (recursively partitioned mixture model for Beta and Gaussian Mixtures) well-
suited for beta-distributed DNA methylation measurements (Houseman et al., 2008). We
!
Figure 3-7: HM450 versus HM27, before and after platform correction. Mean M values
(log2(M/U)) are plotted for HM27 (y axis, 27K) versus HM450 (x axis, 450K), stratified by the
number of CpGs in each probe (top panels left to right: 1-3 CpGs; bottom panels left to right:
4, 5, and 6+ CpGs). Red line indicate a LOESS regression line. a) Tumors, before
correction; b) Technical replicates, before correction; c) Tumors, after correction; d)
Techinical replicates, after correction.
! 104
performed RPMM clustering on the above-mentioned 785 probes for the 373 tumor
samples with beta mixture model. A fanny algorithm (a nonparametric clustering
algorithm) was used for initialization and level-weighted version of Bayesian information
criterion (BIC) as a split criterion for an existing cluster as implemented in the RPMM
package. The clustering result was visualized with a modified version of heatmap.plus,
with samples within each cluster group seriated by hierarchical clustering. The statistical
analysis was done in R.
! 105
Chapter Three Bibliography
American Cancer Society (2013). Cancer Facts & Figures 2013. Atlanta: American
Cancer Society;.
Campan, M., Weisenberger, D.J., Trinh, B., and Laird, P.W. (2009). MethyLight.
Methods in molecular biology 507, 325-337.
Hinoue, T., Weisenberger, D.J., Lange, C.P., Shen, H., Byun, H.M., Van Den Berg, D.,
Malik, S., Pan, F., Noushmehr, H., van Dijk, C.M., et al. (2012). Genome-scale analysis
of aberrant DNA methylation in colorectal cancer. Genome research 22, 271-282.
Houseman, E.A., Christensen, B.C., Yeh, R.F., Marsit, C.J., Karagas, M.R., Wrensch,
M., Nelson, H.H., Wiemels, J., Zheng, S., Wiencke, J.K., et al. (2008). Model-based
clustering of DNA methylation array data: a recursive- partitioning algorithm for high-
dimensional data arising as a mixture of beta distributions. BMC Bioinformatics 9, 365.
Huang, K.C., Rao, P.H., Lau, C.C., Heard, E., Ng, S.K., Brown, C., Mok, S.C., Berkowitz,
R.S., and Ng, S.W. (2002). Relationship of XIST expression and responses of ovarian
cancer to chemotherapy. Mol Cancer Ther 1, 769-776.
Lax, S.F., and Kurman, R.J. (1997). A dualistic model for endometrial carcinogenesis
based on immunohistochemical and molecular genetic analyses. Verhandlungen der
Deutschen Gesellschaft fur Pathologie 81, 228-232.
Noushmehr, H., Weisenberger, D.J., Diefes, K., Phillips, H.S., Pujara, K., Berman, B.P.,
Pan, F., Pelloski, C.E., Sulman, E.P., Bhat, K.P., et al. (2010). Identification of a CpG
island methylator phenotype that defines a distinct subgroup of glioma. Cancer Cell 17,
510-522.
Simpkins, S.B., Bocker, T., Swisher, E.M., Mutch, D.G., Gersell, D.J., Kovatich, A.J.,
Palazzo, J.P., Fishel, R., and Goodfellow, P.J. (1999). MLH1 promoter methylation and
gene silencing is the primary cause of microsatellite instability in sporadic endometrial
cancers. Hum Mol Genet 8, 661-666.
Sirchia, S.M., Ramoscelli, L., Grati, F.R., Barbera, F., Coradini, D., Rossella, F., Porta,
G., Lesma, E., Ruggeri, A., Radice, P., et al. (2005). Loss of the inactive X chromosome
and replication of the active X in BRCA1-defective and wild-type breast cancer cells.
Cancer Res 65, 2139-2146.
! 106
TCGA, C.G.A.R.N. (2008). Comprehensive genomic characterization defines human
glioblastoma genes and core pathways. Nature 455, 1061-1068.
TCGA, C.G.A.R.N. (2011). Integrated genomic analyses of ovarian carcinoma. Nature
474, 609-615.
TCGA, C.G.A.R.N. (2012a). Comprehensive genomic characterization of squamous cell
lung cancers. Nature 489, 519-525.
TCGA, C.G.A.R.N. (2012b). Comprehensive molecular characterization of human colon
and rectal cancer. Nature 487, 330-337.
TCGA, C.G.A.R.N. (2012c). Comprehensive molecular portraits of human breast
tumours. Nature 490, 61-70.
TCGA, C.G.A.R.N. (In Press). Integrated Genomic Analysis of Endometrioid and Serous
Uterine Cancer. Nature.
Toyota, M., Ahuja, N., Ohe-Toyota, M., Herman, J.G., Baylin, S.B., and Issa, J.P. (1999).
CpG island methylator phenotype in colorectal cancer. Proc Natl Acad Sci U S A 96,
8681-8686.
! ! ! ! ! !
! 107
Chapter Four: Clear Cell Renal Cell Carcinoma
Introduction
Kidney cancer accounts for 5% of all cancers in men and 3% in women. Renal
cancer carcinoma (RCC) originating from the proximal convoluted tubule is the most
common type of kidney cancer, responsible for 80% of kidney cancer cases. An
estimated 65,150 new cases of kidney cancer (together with renal pelvis cancer) were
expected to be diagnosed in 2013 (American Cancer Society, 2013). The two common
histological subtypes of RCC, clear cell carcinomas (ccRCC) and papillary renal cell
carcinoma (pRCC), represent 90% of RCCs. ccRCC is the most common type of kidney
cancer (75%) in adults and the most lethal of genitourinary tumors. A genetic hallmark of
ccRCC is chromosome 3p deletion (Zbar et al., 1987), and ccRCC and is one of the few
tumors that are known to be tightly associated with a specific gene. Germline mutations
in the von Hippel-Lindau (VHL) gene located on 3p are tightly associated with hereditary
ccRCCs (Zbar et al., 1996), and somatic VHL mutations explain for about 33-57% of
ccRCCs (Brauch et al., 2000). Existing literatures also report on VHL promoter
hypermethylation in 20% of the cases (Herman et al., 1994). It is noteworthy that VHL
inactivation is not associated with pRCC, which is characterized by genetic abnormalities
in the MET and PRCC genes.
Two other genes located on chromosome 3p, both being epigenetic regulators,
have recently been linked to ccRCC. The polybromo-containing PBRM1 in the PBAF
SWI/SNF complex was recently found to be the second most mutated gene, with a
mutation frequency of 41%, in clear cell renal cell carcinomas (Varela et al., 2011).
BAP1, also from 3p, is mutated in 15% of ccRCCs (Pena-Llopis et al., 2012). In addition,
the non-redundant H3K36 methyltransferase SETD2 is mutated in 12% of ccRCCs
! ! ! ! ! !
! 108
(Dalgliesh et al., 2010).
A recent study reconstructed the phylogenetic structure of molecular events in
ccRCC with multiple spatially separated samples from the same tumors (Gerlinger et al.,
2012). In both of the two patients studied, distinct SETD2 inactivating mutations were
found in different parts of the same tumor. Immunohistochemistry staining confirmed
H3K36me3 loss in all the mutant tumors. This convergent somatic evolution indicates
that failure to establish H3K36 methylation marks provides a strong selective advantage
relatively late in ccRCC progression. A similar molecular convergence was found for
KDM5C, an H3K4 demethylase, in one of the two patients. This, together with recurrent
mutations in other epigenetic regulators, shows that epigenetic dysregulation, often
mediated by genetic events, is important in advanced ccRCCs.
! ! ! ! ! !
! 109
Results
Overall DNA Methylation Characteristics of ccRCC
Overall, no strong DNA methylation subtypes for ccRCCs are observed in an
unsupervised analysis, after excluding possible chromophobe kidney samples from the
study cohort (Figure 4-1). The clusters made by the RPMM algorithm do not exhibit
significant difference in their DNA methylation profiles, but rather only slight difference at
CpG poor regions (Figure 4-1). Tumor purity as well as tumor stage/grade explain a
great portion of the variation in the DNA methylation pattern observed. Increasing
promoter hypermethylation frequency correlated with higher stage and grade (Figure 4-
2), indicating that ccRCCs likely represent a single disease entity with increased
!
Figure 4-1: Unsupervised clustering of DNA methylation data for 373 tumors.
Clustering on 582 most variable CpG loci gave four clusters (top color bar labeled ‘rpmm’),
but visual examinations shows that the subtypes are not well-defined. The
heterogeneity is captured well with stage/grade as well as purity (top color bars).
!
! ! ! ! ! !
! 110
molecular alteration during disease development and progression. ccRCC tumors have
extensive leukocyte infiltration, and such infiltration increases with tumor stage/grade.
Epigenetic Silencing of Genes
We observed epigenetic silencing of 290 genes in at least 5% tumors (Figure 4-
3). These 290 genes include VHL, silenced in about 7% of ccRCC tumors, which was
mutually exclusive with mutation of VHL, reflecting the central role of this locus in
!
Figure 4-2: Promoter DNA hypermethylation increases with tumor stage/grade (a-b) Overall
promoter DNA hypermethylation frequency in the tumor increases with rising stage (a) and
grade (b). The promoter DNA hypermethylation frequency is calculated as the percentage of
CpG loci hypermethylated among 15,101 loci, which are unmethylated in the normal kidney
tissue and normal white blood cells (supplementary methods). Notches in the boxplots
indicate 95% confidence interval of the medians. (c) Heatmap visualizing the DNA
methylation across the 1,087 CpG loci that hypermethylated in >5% of the tumors. The
tumors are sorted by average DNA methylation level across the 1,087 loci. The stage and
grade of the tumor are indicated on the top with color bars with light to dark green indicating
low stage to high stage and light red to dark red indicating low grade to high grade.
!
! ! ! ! ! !
! 111
ccRCC.! This percentage is lower than the ~20% frequency which has been reported
previously for ccRCC cell lines (Herman et al., 1994), possibly because of reduced
sensitivity of detection caused by contaminating non-malignant cells in the current cohort
of primary tumors. Indeed, Herman et al observed this for primary tumors as well, and it
has been reported that renal cell carcinomas could contain up to 60% lymphocytes (Zbar
et al., 1987). In our analysis we also observed high degree of leukocyte infiltration in the
tumors.
!
Figure 4-3: 290 epigenetically silenced genes in ccRCCs. Width of bars indicates the
frequency of epigenetic silencing for each gene. VHL is highlighted in red.
!
! ! ! ! ! !
! 112
The top-ranked gene by inverse correlation between gene expression and DNA
methylation was UQCRH, hypermethylated in 36% of the tumors (Figure 4-3; Figure 4-
5). UQCRH has been previously suggested to be a tumor suppressor (Modena et al.,
2003), but not linked to ccRCC. Similarly, the protocadherin (PCDH) family members
are often silenced in ccRCCs too. The long-range epigenetic silencing of the PCDHs
was reported for Wilms’ Tumors (Dallosso et al., 2009).!PCDH10 silencing was reported
for different cancers not including clear cell kidney and it was postulated that they might
be tumor suppressor genes by suppressing the canonical Wnt pathway (Ying et al.,
2006).! CCND2 encoding Cyclin D2, although intuitively deemed oncogenic, is also
!
Figure 4-4: VHL epigenetic silencing. The mRNA expression level (y axis, RNA-seq
RPKM) versus DNA methylation level (x axis, beta value) for VHL is plotted for each tumor.
VHL mutation status is indicated for each sample with a black or red color. Mean values (solid
lines) and normal range (dashed lines) for DNA methylation and gene expression level in the
normal kidney samples are also indicated. VHL epigenetic silencing in a subset of the tumors
is evident with high DNA methylation level and low expression level, and is mutually exclusive
with VHL mutations found.!
! ! ! ! ! !
! 113
epigenetically silenced in ccRCCs. It was reported that epigenetic silencing of CCND2 in
prostate cancer is associated with bad prognosis (Padar et al., 2003).
The RAS association domain family 1A gene (RASSF1A), has been reported to
be epigenetically silenced in ccRCCs (Dreijerink et al., 2001; Morrissey et al., 2001;
Peters et al., 2007; Yoon et al., 2001). However, in our study, it is not identified as an
epigenetically silenced gene, as it is methylated in all of the normal kidneys. The fact
that it is methylated in all of the normal kidneys, and not only tumors with RASSF1A
methylation, argues against proposal that the methylation observed in normal kidney
might be attributable to field effects (Peters et al., 2007)., In fact, about 1/6 of the ccRCC
tumors exhibit loss of methylation at the RASSF1A promoter. Indeed, the tumors have
higher RASSF1A expression compared to the normal samples.! ! Therefore, the body of
!
Figure 4-5: UQCRH epigenetic silencing. y axis: Gene expression, as measured by
RNA-seq RPKM; x axis: DNA methylation, as measured by beta value at cg21576698.
Each dot is a sample, with green indicating normal and black indicating tumors.
!
! ! ! ! ! !
! 114
literature on cancer-associated RASSF1A epigenetic silencing should be re-evaluated in
comparison to the associated normal tissue.
DNA Methylation Profile Associated with SETD2 Mutation
We also evaluated the global consequences of mutation in specific epigenetic
modifiers. Mutations in SETD2, a non-redundant H3K36 methyltransferase, were
associated with increased loss of DNA methylation at non-promoter regions (Figure 4-6).
This discovery is consistent with the emerging view that H3K36 trimethylation may be
involved in the maintenance of a heterochromatic state (Wagner and Carpenter, 2012),
whereby DNA methyltransferase 3A (DNMT3A) binds H3K36me3 and methylates
nearby DNA (Dhayalan et al., 2010). Thus, reductions of H3K36me3 through SETD2
inactivation could lead indirectly to regional loss of DNA methylation.
!
Figure 4-6: RASSF1A is not epigenetically silenced in ccRCCs compared to the
normals. Trancripts annotated to the RASSF1 gene region is plotted on the left, with
CpG islands associated with promoter 1A (top) and 1C (bottom) highlighted with boxes.
Each tick in the box indicate a CpG locus interrogated by an HM450 probe. HM450 has
good coverage of both promoters. Promoter 1A is methylated in all normal, and therefore
the ‘epigenetic silencing’ reported for ccRCCs in earlier literature may not be cancer-
specific. On the contrary, about 1/6 of the tumors have loss of methylation at RASSF1A
compared to the normal.
!
! ! ! ! ! !
! 115
To further investigate the significance of the SETD2-associated DNA methylation
changes, we permuted the SETD2 labels 1,000 times, and plotted the distribution of the
number of significant genes, using the same FDR and effect size cutoffs (Figure 4-8).
The vast majority (779) of the 1,000 permutations did not yield any loci with significant
DNA methylation changes, with an average of 0.494 significant probes for each
permutation. The highest number of significant loci in any of the permutations was 33
loci. In contrast, we observed 2,557 loci with significant DNA methylation changes
associated with SETD2 mutation with the correct labels. We conclude that the difference
!
Figure 4-7: A distinct DNA methylation profile associated with SETD2 mutation. (a)
Volcano plots showing a comparison of DNA methylation for SETD2 mutant versus non-
mutant tumors, done on 224 samples assayed on the HumanMethylation450 platform. The
y-axis is the p- value corrected for multiple comparisons. The area not shaded indicates the
CpG loci plotted in Panel D (Benjamini-Hochberg FDR=0.001, difference in mean beta value
>0.1; n=2,557). (b) Heatmap showing the CpG loci with DNA methylation associated with
SETD2 mutation. The loci are split into loci hypomethylated in SETD2 mutants (top panel;
n=1,251), and those hypermethylated in SETD2 mutants (bottom panel; n=1,306) compared
to non-mutants. Top color bars indicate the SETD2 mRNA expression [green to red indicate
low to high log2(RPKM)], and SETD2 mutation status in the tumors, respectively. A
spectrum from blue to red indicates low to high DNA methylation (a beta value of 0 to 1). A
gray-scale row-side color bar on the left-hand side shows the relative number of reads
overlapping each of the loci, in an H3K36me3 ChIP-seq experiment in normal adult kidney
(http://nihroadmap.nih.gov/epigenomics/). White to black corresponds to low (0) to high (34)
read count. 14 normal kidney samples with both expression and DNA methylation data are
also plotted on the left. Among the tumors without SETD2 mutations, six (arrowhead) have
both the signature pattern of SETD2 mutation and low SETD2 mRNA expression.
!
! ! ! ! ! !
! 116
observed between SETD2 mutants and wildtype tumors is unlikely to be attributable to
chance.
!
Figure 4-8: Permutation of the SETD2 label.
Distribution of the number of significant loci after
permuting the SETD2 labels for 1000 times, compared to
2,557 loci with the correct label.
!
! ! ! ! ! !
! 117
Methods
Array-based DNA methylation assay
We used two Illumina Infinium DNA methylation platforms, HumanMethylation27
(HM27) BeadChip and HumanMethylation450 (HM450) BeadChip (Illumina, San Diego,
CA) to obtain DNA methylation profiles of 502 TCGA clear cell renal carcinoma samples
and 359 adjacent non-tumor kidney tissue samples. Twelve technical replicates were
also included in the assay to monitor technical variations, with six on the HM27 platform
and six on the HM450 platform. We included 444 of the tumor samples and all normal
samples in the ‘extended’ list used for analyses based on DNA methylation data only,
and 373 of the 444 tumor samples in the 'core' list used for cross-platform comparisons.
The Infinium HM27 array targets 27,578 CpG sites located in proximity to the
transcription start sites of 14,475 consensus coding sequencing (CCDS) in the NCBI
Database (Genome Build 36). The Infinium HM450 array targets 482,421 CpG sites
through out the genome and covers 99% of RefSeq genes. It covers 96% of CpG
islands, with additional coverage in island shores and the regions flanking them. The
assay probe sequences and information for each interrogated CpG site on both Infinium
DNA methylation platforms can be found in the MAGE-TAB ADF (Array Design Format)
file available through the TCGA Data Portal (http://tcga-data.nci.nih.gov/tcga/).
We performed bisulfite conversion on 1 µg of genomic DNA from each sample
using the EZ-96 DNA Methylation Kit (Zymo Research, Irvine, CA) according to the
manufacturer’s instructions. We assessed the amount of bisulfite converted DNA and
completeness of bisulfite conversion using a panel of MethyLight-based quality control
(QC) reactions as described previously (Campan et al., 2009). All the TCGA samples
passed our QC tests and entered the Infinium DNA methylation assay pipeline.
! ! ! ! ! !
! 118
Bisulfite-converted DNA was whole genome amplified (WGA) and enzymatically
fragmented prior to hybridization to the arrays. BeadArrays were scanned using the
Illumina iScan technology, and the IDAT files (Level 1 data) were used to extract the
intensities (Level 2 data) and calculate the beta value (Level 3 data) for each probe and
sample with the R-based methylumi package.
The level of DNA methylation at each CpG locus is summarized as a beta (β)
value calculated as (M/(M+U)), ranging from 0 to 1, which represents the the ratio of the
methylated probe intensity to the overall intensity at each CpG locus. A p-value
comparing the intensity for each probe to the background level was also calculated with
the methylumi package, and data points with a detection p-value >0.05 were deemed
not significantly different from background measurements, and therefore were masked
as “NA” in the Level 2 and 3 in HM27 and Level 3 in HM450 data packages, as detailed
below.
TCGA data packages
The three data levels are described below and are present on the TCGA Data
Portal website (http://tcga-data.nci.nih.gov/tcga/). Please note that with continuing
updates of genomic databases, data archive revisions become available at the TCGA
Data Portal.
HM27: Level 1 - Level 1 data packages contain the non-background corrected
signal intensities of the M and U probes and the mean negative control cy5 (red) and cy3
(green) signal intensities. A detection p-value for each data point, the number of
replicate beads for M and U probes as well as the standard error of M, U, and control
probe signal intensities are also provided. It is important to note that for some CpG
! ! ! ! ! !
! 119
targets, both M and U measurements will be cy3, and for others both will be cy5. To
resolve ambiguities regarding this subtlety of the Infinium DNA Methylation assay, we
have labeled the cy3 and cy5 values deposited to the DCC as “Methylated Signal
Intensity” and “Unmethylated Signal Intensity”. The information of the color channel for
each CpG locus is contained in the MAGE-TAB ADF file deposited in the DCC. Level 2 -
Level 2 data files contain the β-value calculations for each probe and sample. Data
points with detection p-values >0.05 were not considered to be significantly different
from background, and were masked as “NA”. Level 3 - Level 3 data contain β-value
calculations, HUGO gene symbol, chromosome number and genomic coordinate for
each targeted CpG site on the array. In addition, we masked data points with "NA" from
the probes that 1) contain known single nucleotide polymorphisms (SNPs) after
comparison to the dbSNP database (Build 132), 2) contain repetitive sequence elements
that cover the targeted CpG locus in each 50 bp probe sequence, 3) are not uniquely
aligned to the human genome (NCBI build 36.1) at 20 nucleotides at the 3’ terminus of
the probe sequence, 4) span known regions of small insertions and deletions (indels) in
the human genome (dbSNP build 130).
HM450: Level 1 - Level 1 data contain raw IDAT files. IDAT files are the direct
output from the scanning program. Level 2 - Level 2 data contain background corrected
signal intensities of the M and U probes. Level 3 - Level 3 data files contain β-value
calculations and masked data points with "NA" from the probes that are annotated as
having a SNP within 10 base pairs of the interrogated locus (HM27 carryover or recently
discovered). The genomic characteristics for each probe are available for download via
Illumina (www.illumina.com).
! ! ! ! ! !
! 120
The following data archives were used for the analyses described in this
manuscript:
[1] "jhu-usc.edu_KIRC.HumanMethylation27.Level_3.1.3.0"
[2] "jhu-usc.edu_KIRC.HumanMethylation27.Level_3.2.3.0"
[3] "jhu-usc.edu_KIRC.HumanMethylation27.Level_3.3.3.0"
[4] "jhu-usc.edu_KIRC.HumanMethylation27.Level_3.4.3.0"
[5] "jhu-usc.edu_KIRC.HumanMethylation27.Level_3.5.3.0"
[6] "jhu-usc.edu_KIRC.HumanMethylation450.Level_3.1.4.0"
[7] "jhu-usc.edu_KIRC.HumanMethylation450.Level_3.2.4.0"
[8] "jhu-usc.edu_KIRC.HumanMethylation450.Level_3.3.4.0"
[9] "jhu-usc.edu_KIRC.HumanMethylation450.Level_3.4.4.0"
[10] "jhu-usc.edu_KIRC.HumanMethylation450.Level_3.5.4.0"
[11] "jhu-usc.edu_KIRC.HumanMethylation450.Level_3.6.4.0"
Merging HM27 and HM450 Data
The shared probe set between HM27 and HM450 platforms (n=25,978) were
used for the analysis. Out of the 25,978 probes, 887 probes were masked due to
detection p-value, repeats and SNPs and non-uniquely mapped probes (n=25,091
remaining). We observed batch and platform specific effects with the technical
replicates. To alleviate systematic platform-specific effects (dye bias, background level,
etc.) we fitted a LOESS regression model between the two platforms using the twelve
technical replicates. We normalized the HM450 data against the HM27 data with this
fitted model on the M values, stratified by the number of CpGs in the probe
(CpG=1,2,3,4,5,6+). M value is the log 2 ratio of Methylated (M) intensity and
Unmethylated (U) intensity and better satisfies the linearity assumption. The M values
! ! ! ! ! !
! 121
were then transformed back to beta values with the equation Beta=2^M/(2^M+1). In
order to further remove probes that were not fitted well with the LOESS model, as well
as probes prone to technical variation other than platform-specific effects, we calculated
probe-wise standard deviation across the twelve technical replicates and masked any
probe with a probe-wise standard deviation of greater than 0.1 (n=24,383 remaining).
Global Hypermethylation and Clinical Stage/Grade
We investigated all CpG loci (n=16,123) assayed on both HM27 and HM450
platforms that are unmethylated in the normal adjacent kidney tissue (average DNA
methylation beta value <0.2) for cancer-specific hypermethylation. We further excluded
1,021 loci methylated in the normal white blood cells (average DNA methylation beta
value > 0.2) to avoid ‘passive’ hypermethylation signature due to blood contamination or
lymphocyte infiltration at those loci. We calculated the percentage of hypermethylated
(beta value>0.2) loci. Boxplots of this percentage for normal and tumors of different
stages were used to visualize the trend of increasing DNA hypermethylation with
advancing stages and grade.
Epigenetic Silencing Calls
Again, only intersect probes (n=25,091) of HM27 and HM450 were used for this
analysis. For each gene, we chose DNA methylation probes that satisfy the following
criteria:
1. The locus studied should be unmethylated in the normal kidney tissue: 95th
percentile for methylation in normals <0.2; we use the 95
th
percentile instead of the
maximum to allow for field effects in 5% of the normal;
2. DNA methylation at the locus studied should be inversely correlated with expression
level of the gene: correlation coefficient with log2(RPKM+1) < -0.2, and adjusted p-value
! ! ! ! ! !
! 122
testing for correlation < 0.005;
3. The locus studied should be methylated in some of the tumors: 95th percentile for
DNA methylation beta value in tumor > 0.2;
4. The hypermethylation level in the tumor should be considerable: maximum DNA
methylation beta value in the tumor > 0.5.
Any gene with at least one such CpG locus detected was called epigenetic
silenced. Then, for each sample, we looked at all probes that satisfy the above criteria
for each silenced gene. A sample is called silencing if it satisfies the following criteria:
1. Overall hypermethylation: the mean methylation across those loci > 0.2;
2. Consistency across various loci: DNA methylation at each CpG locus uniformly >
95th quantile (normal).
DNA Methylation Pattern Associated with SETD2 mutation
We used a univariate two-sample t-test to evaluate whether DNA methylation
level at each CpG locus investigated at the HM450 platform was different in the SETD2
mutants (n=32) and wildtype tumors (n=192). The p-values were then corrected for
multiple comparisons using the Benjamini-Hochberg procedure. The adjusted p-value
was plotted against the difference between the mean beta value in SETD2 mutants and
mean beta value in SETD2 wildtype tumors (volcano plot). A heatmap was used to
visualize a subset of the loci (absolute difference in beta value > 0.1, adjusted p-value
<0.001) at which SETD2 mutants were significantly differently methylated from the
wildtype tumors. Roadmap (http://nihroadmap.nih.gov/epigenomics/) human adult kidney
H3K36me3 ChIP-Seq data were downloaded from GEO (Accession: GSM773000) and
the number of reads overlapping with each of the loci in the heatmap was plotted as a
rowside color bar.
! ! ! ! ! !
! 123
Statistics. All statistical analyses were conducted in R version 2.15.0 (2012-03-30). All
p-values reported were two-sided.
! !
! ! ! ! ! !
! 124
Chapter Four Bibliography:
American Cancer Society (2013). Cancer Facts & Figures 2013. Atlanta: American
Cancer Society;.
Brauch, H., Weirich, G., Brieger, J., Glavac, D., Rodl, H., Eichinger, M., Feurer, M.,
Weidt, E., Puranakanitstha, C., Neuhaus, C., et al. (2000). VHL alterations in human
clear cell renal cell carcinoma: association with advanced tumor stage and a novel hot
spot mutation. Cancer Res 60, 1942-1948.
Campan, M., Weisenberger, D.J., Trinh, B., and Laird, P.W. (2009). MethyLight.
Methods in molecular biology 507, 325-337.
Dalgliesh, G.L., Furge, K., Greenman, C., Chen, L., Bignell, G., Butler, A., Davies, H.,
Edkins, S., Hardy, C., Latimer, C., et al. (2010). Systematic sequencing of renal
carcinoma reveals inactivation of histone modifying genes. Nature 463, 360-363.
Dallosso, A.R., Hancock, A.L., Szemes, M., Moorwood, K., Chilukamarri, L., Tsai, H.H.,
Sarkar, A., Barasch, J., Vuononvirta, R., Jones, C., et al. (2009). Frequent long-range
epigenetic silencing of protocadherin gene clusters on chromosome 5q31 in Wilms'
tumor. PLoS Genet 5, e1000745.
Dhayalan, A., Rajavelu, A., Rathert, P., Tamas, R., Jurkowska, R.Z., Ragozin, S., and
Jeltsch, A. (2010). The Dnmt3a PWWP domain reads histone 3 lysine 36 trimethylation
and guides DNA methylation. J Biol Chem 285, 26114-26120.
Dreijerink, K., Braga, E., Kuzmin, I., Geil, L., Duh, F.M., Angeloni, D., Zbar, B., Lerman,
M.I., Stanbridge, E.J., Minna, J.D., et al. (2001). The candidate tumor suppressor gene,
RASSF1A, from human chromosome 3p21.3 is involved in kidney tumorigenesis. Proc
Natl Acad Sci U S A 98, 7504-7509.
Gerlinger, M., Rowan, A.J., Horswell, S., Larkin, J., Endesfelder, D., Gronroos, E.,
Martinez, P., Matthews, N., Stewart, A., Tarpey, P., et al. (2012). Intratumor
heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J
Med 366, 883-892.
Herman, J.G., Latif, F., Weng, Y., Lerman, M.I., Zbar, B., Liu, S., Samid, D., Duan, D.S.,
Gnarra, J.R., Linehan, W.M., et al. (1994). Silencing of the VHL tumor-suppressor gene
by DNA methylation in renal carcinoma. Proc Natl Acad Sci U S A 91, 9700-9704.
! ! ! ! ! !
! 125
Modena, P., Testi, M.A., Facchinetti, F., Mezzanzanica, D., Radice, M.T., Pilotti, S., and
Sozzi, G. (2003). UQCRH gene encoding mitochondrial Hinge protein is interrupted by a
translocation in a soft-tissue sarcoma and epigenetically inactivated in some cancer cell
lines. Oncogene 22, 4586-4593.
Morrissey, C., Martinez, A., Zatyka, M., Agathanggelou, A., Honorio, S., Astuti, D.,
Morgan, N.V., Moch, H., Richards, F.M., Kishida, T., et al. (2001). Epigenetic inactivation
of the RASSF1A 3p21.3 tumor suppressor gene in both clear cell and papillary renal cell
carcinoma. Cancer Res 61, 7277-7281.
Padar, A., Sathyanarayana, U.G., Suzuki, M., Maruyama, R., Hsieh, J.T., Frenkel, E.P.,
Minna, J.D., and Gazdar, A.F. (2003). Inactivation of cyclin D2 gene in prostate cancers
by aberrant promoter methylation. Clin Cancer Res 9, 4730-4734.
Pena-Llopis, S., Vega-Rubin-de-Celis, S., Liao, A., Leng, N., Pavia-Jimenez, A., Wang,
S., Yamasaki, T., Zhrebker, L., Sivanand, S., Spence, P., et al. (2012). BAP1 loss
defines a new class of renal cell carcinoma. Nat Genet 44, 751-759.
Peters, I., Rehmet, K., Wilke, N., Kuczyk, M.A., Hennenlotter, J., Eilers, T., Machtens,
S., Jonas, U., and Serth, J. (2007). RASSF1A promoter methylation and expression
analysis in normal and neoplastic kidney indicates a role in early tumorigenesis. Mol
Cancer 6, 49.
Varela, I., Tarpey, P., Raine, K., Huang, D., Ong, C.K., Stephens, P., Davies, H., Jones,
D., Lin, M.L., Teague, J., et al. (2011). Exome sequencing identifies frequent mutation of
the SWI/SNF complex gene PBRM1 in renal carcinoma. Nature 469, 539-542.
Wagner, E.J., and Carpenter, P.B. (2012). Understanding the language of Lys36
methylation at histone H3. Nature reviews Molecular cell biology 13, 115-126.
Ying, J., Li, H., Seng, T.J., Langford, C., Srivastava, G., Tsao, S.W., Putti, T., Murray, P.,
Chan, A.T., and Tao, Q. (2006). Functional epigenetics identifies a protocadherin
PCDH10 as a candidate tumor suppressor for nasopharyngeal, esophageal and multiple
other carcinomas with frequent methylation. Oncogene 25, 1070-1080.
Yoon, J.H., Dammann, R., and Pfeifer, G.P. (2001). Hypermethylation of the CpG island
of the RASSF1A gene in ovarian and renal cell carcinomas. Int J Cancer 94, 212-217.
! ! ! ! ! !
! 126
Zbar, B., Brauch, H., Talmadge, C., and Linehan, M. (1987). Loss of alleles of loci on the
short arm of chromosome 3 in renal cell carcinoma. Nature 327, 721-724.
Zbar, B., Kishida, T., Chen, F., Schmidt, L., Maher, E.R., Richards, F.M., Crossey, P.A.,
Webster, A.R., Affara, N.A., Ferguson-Smith, M.A., et al. (1996). Germline mutations in
the Von Hippel-Lindau disease (VHL) gene in families from North America, Europe, and
Japan. Hum Mutat 8, 348-357.
127
!
Chapter Five: Epigenetic Analysis Leads to Identification of
HNF1B as a Subtype-Specific Susceptibility Gene
for Ovarian Cancer
Introduction
Invasive epithelial ovarian cancer (EOC) has a strong heritable component
(Lichtenstein et al., 2000), with an approximate three-fold increased risk associated with
a first-degree family history (Auranen et al., 1996). Much of the excess familial risk
observed for EOC is unexplained (Antoniou and Easton, 2006), and efforts to identify
common susceptibility genes have proved difficult. Seven regions harboring
susceptibility single nucleotide polymorphisms (SNPs) for ovarian cancer have been
identified through genome-wide association studies (Bolton et al., 2010; Goode et al.,
2010; Pharaoh et al., In Press; Song et al., 2009) thus far, but candidate gene studies
have been largely unsuccessful (Bolton et al., 2012).
The Cancer Genome Atlas (TCGA) has fully characterized more than 500 serous
EOC cases with respect to somatic mutation, DNA methylation, mRNA expression and
germline genetic variants (TCGA, 2011). These data are publicly available and can be
analyzed to identify candidate genes for association studies of the disease.
We conducted such an analysis of TCGA data (Chapter Two) and found a unique
expression and methylation pattern of HNF1B characterized by down-regulation of
expression in most cases, with epigenetic silencing in about half of the cases,
suggesting it might play a role in the serous subtype of ovarian cancer (Appendix A). In
contrast, HNF1B overexpression is common in clear cell ovarian cancer (Tsuchiya et al.,
128
!
2003). The HNF1B gene (formerly known as TCF2) encodes a POU-domain containing
a tissue-specific transcription factor, and mutations in the gene cause maturity onset
diabetes of the young type 5 (Horikawa et al., 1997). HNF1B is also a susceptibility gene
for type II diabetes (Gudmundsson et al., 2007; Winckler et al., 2007), prostate cancer
(Berndt et al., 2011; Gudmundsson et al., 2007; Sun et al., 2008; Thomas et al., 2008)
and uterine cancer (Spurdle et al., 2011).
Therefore, we characterized this gene in ovarian cancer in a comprehensive way
and found evidence of a differential effect of HNF1B on the serous and clear cell
subtypes of ovarian cancer. It appears that HNF1B plays a loss-of-function role in serous
and a gain-of-function role in clear cell ovarian cancers and variants in this gene
differentially affect genetic susceptibility to these subtypes.
129
!
Results
DNA Methylation/Expression Analysis
From TCGA data (Chapter Two), HNF1B was observed to be epigenetically
silenced in approximately half of the 576 primary serous ovarian tumors and down-
regulated by another mechanism in most of the other tumors, whereas no evidence of
methylation was seen in the normal fallopian tube samples (Figure 5-1a) available from
TCGA. This suggested that HNF1B inactivation is important for serous ovarian cancer.
We further assessed HNF1B promoter methylation in an independent dataset (OCRF;
see Methods below) and found the promoter region to be methylated in 42% of serous
tumors and in none of the clear cell ovarian tumors (Figure 1b). This unique pattern in
serous tumors in contrast to clear cell cancers led to the evaluation of HNF1B as a
candidate subtype-specific susceptibility gene for ovarian cancer.
130
!
b.
DNA METHYLATION
Beta Value
TCGA Normal
TCGA Serous
USC Serous
USC Clear Cell
a.
DNA METHYLATION
Beta Value
mRNA EXPRESSION
Log Ratio
c. d. e.
Figure 5-1: Identification of HNF1B as a subtype-specific candidate gene for ovarian
cancer and establishment of it as a susceptibility gene. (a) mRNA expression (y-axis)
versus DNA methylation (x-axis) in serous ovarian tumors. Each blue dot is a serous tumor
sample, while each pink dot is one of ten normal fallopian tube samples. HNF1B is silenced in
the majority of these tumors, either by an epigenetic (bottom right, high DNA methylation and
low mRNA expression) or an unknown alternative mechanism. (b) HNF1B promoter DNA
methylation differs by histological subtype. While unmethylated in the normal fallopian tissue,
this locus is hypermethylated (beta value >0.2) in approximately 50% of the TCGA (n=576)
serous cases as well as another independent set of 32 serous tumor samples (OCRF panel),
but remains unmethylated in clear cell tumors (OCRF panel) (n=4). These data are consistent
with reported HNF1B expression in the clear cell tumors. (c-e): The mRNA expression data
from (a) were integrated from three platforms, and we observe the same pattern with each
individual expression platform:! (c) Affymetrix Human Exon 1.0 ST array, 528 tumors and 10
normal fallopian tube samples; (d) Affymetrix U133A, 568 tumors and eight normal fallopian
tube samples; (e) Agilent G4502, 543 tumors and four normals. Plotted as the y-axes are log
intensities for the single-color-channel arrays (HuEx1.0 and Affymetrix) and log ratios for the
two-color-channel array (Agilent); and the x-axes indicate beta values for DNA methylation from
0 (unmethylated) to 1 (methylated).
131
!
SNP Analysis
With all invasive cancer subtypes considered together, we found no genome-
wide significant (P<5x10
-8
) HNF1B SNP associations among women of European
ancestry (Table 5-1). However, when analyses were stratified by histologic subtype, we
observed genome-wide significant results for both serous and clear cell EOC subtypes,
but with risk associations in opposite directions. The association was similar for high-
and low-grade serous cancers. There was no evidence of association for mucinous or
chr17
Chromosome 17
Axis
36.04 mb
36.05 mb
36.06 mb
36.07 mb
36.08 mb
36.09 mb
36.1 mb
36.11 mb
36.12 mb
36.13 mb
36.14 mb
36.15 mb
36.16 mb
36.17 mb
0
2
4
6
8
10
−log10(p)
Serous
● ●●
● ●
● ●
●
● ● ●
● ● ●
● ● ●
● ● ● ●
● ●
● ●
●
●
●
●● ●
● ● ●
●
● ● ●
●
●
●
●
●
● ● ●● ●
● ●
●
●
●
●
●
●
●
● ●
●
●
●●
●
● ●
●
●
●●
●● ● ●
●
●
●
●
●
●
● ● ●
● ●●
●●
●
●
● ● ●
●
● ● ●
●
●
●
● ●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
● ● ●
●
●● ●
●
●●
● ●●
●
● ● ●
● ●
●
●
● ●
●
● ●
● ● ● ●
●
● ●
●
● ● ●
●
● ● ● ● ●●
●
●
●
●
● ● ● ● ●
●
●
●●
●
● ● ● ●
●
●
●
●
● ●
●
●
● ●●● ● ● ●●●
●
● ● ● ● ● ● ●
●
●● ●● ● ● ●
●
●
●
●
● ● ●●
●
●
●●
●
● ● ●●
● ●
●
●●
●
●
●
●● ●
●
●
●
●
● ● ●●● ●
●
● ● ●
●
●
● ●
●
●
●
●
●
●
● ●
● ●
●
●
● ●
●
● ● ● ●
●
●
●
●
● ●
●
● ●
●
● ● ●
●
● ●
●
●
●
●
●
● ● ●
●
●
● ●
●
●
●
● ●
●● ● ● ● ● ● ●
● ●
● ● ●
●
●
●
●●
● ●
●
●
●
●
●
●
● ●
●●
●●
●
●
●
●
● ● ●
● ● ●
●
●● ●
●
● ● ●
●
●
● ●
●
●
●
●●
●
●
● ●
● ●
● ● ● ●
●
●
● ●●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
● ● ●
●
●
●
●
●
●
●
●
●●
●
●
● ●
●
●
●
●
●
● ●
●
● ●
●
●
●
●
● ●
●
●
●
●
●
●
●
● ●
●
●
●
●
●
●
● ●
● ● ● ●
●
● ●
●
●●
● ●
●● ● ● ● ● ●
●
●
● ● ● ●
●
●
●
●
●
0
2
4
6
8
10
−log10(p)
Endometrioid
● ●● ● ●● ●● ● ● ● ● ● ●● ● ●●● ● ●
●
● ● ● ● ●
● ●● ●● ●
● ●
●
● ● ● ●
● ●
● ● ●
●● ●
● ● ● ● ● ● ●
●
●
● ● ● ● ●● ● ● ● ● ●
●●
● ● ● ● ● ● ● ● ●●●
● ●
●
● ●
●●● ● ● ●●
●
●
● ●
●
● ●
● ● ●●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
● ●
●
● ●
●
●
● ● ●
● ●●
● ●●
● ● ● ● ● ●
●
● ● ●● ● ● ● ● ● ● ● ● ● ●
●
●
● ● ● ●
● ● ●● ●● ●
●
●
● ● ●
●
●
●
●
● ●
● ●
● ●
●
●
●
●
●
● ●
●
●
●
●
●
●
● ●
●
● ●
●
●
●
●
●
●
●
● ●●
●
● ●
● ●
●
● ● ●
● ● ●●
●
●
●
●
●
●
● ●●
● ●
●
●●
●
●
●
● ● ●●● ●● ● ● ●●● ●
●
● ● ● ● ● ●● ●
●
●
●
●● ● ● ● ●
●
●
● ●
●
● ● ● ● ●
● ● ●
● ●
●
● ● ● ● ● ● ● ● ●
●
●
●
●
●
● ●● ● ● ● ● ●
●
●● ● ●
●
● ●
● ●
●
●● ● ● ● ● ●●
●
●
●
●
●
●
● ●
●
●
● ● ●
●
● ●
●● ● ● ●● ● ● ● ●
●
● ● ●
●
● ● ● ●● ●● ●
● ●
●
●● ●●● ●● ● ● ● ● ●
●
●
● ● ●
●
● ●
● ●
●● ●● ●●
● ●
●
● ●
●
●
●
●
●
●
●
●
●
● ●
●
●
● ●
●
●●
● ● ● ● ●
●
●
●
● ●
●●
●
● ●
●
●
●
●
● ●
●● ● ●
● ● ● ●
●
● ● ● ●
●
●
●
●
● ●
● ●
● ●
● ●●
●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
●
● ●
0
2
4
6
8
10
−log10(p)
Mucinous
●
●● ● ●● ●● ● ●
● ● ● ●
●
●
●
●● ● ● ● ● ●
● ●
●
● ●● ●●
●●
● ●
● ●
● ●
●
● ● ● ● ●● ● ● ●
● ●
● ● ●
●
●
● ● ● ● ●● ● ● ● ●
●
●● ●
● ● ● ● ● ● ● ●
●
● ●
●
● ●●
●●
●
●
● ●●
●
●
● ●
● ● ●● ● ●●
●
● ●
●
●
●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
● ●
●
●
●● ●
●
●●
● ●●
●
● ● ●
● ●
●
●
● ●
●
● ●
● ● ● ●
●
● ●
●
●
●
●
●
● ●
● ● ●●
●
●
●
●
● ● ● ● ●
●
●
●
●
●
● ●
● ● ●
● ● ●
●
●
●
● ●
●
●● ●
●
●
●
●
● ● ● ●
●
● ● ●●
●
● ●
●
● ● ●
●
●
● ●
● ● ●●
●
●
●
●
●
●
●
●●
● ●
●
●●
●
●
●
● ● ●
●
●
●
● ● ● ●●● ● ●● ● ● ●
●
●
● ● ●
●
●
● ●
● ● ● ● ● ●
● ●
●
● ● ● ●
● ●
●
● ● ● ● ● ● ● ● ● ● ●
● ● ●
●
● ● ●
● ●● ●
● ●
●
● ● ● ● ●
● ●
●
●
●
●
●
●
●
●
● ●
●
●●
●
●
● ●
●
● ● ● ●
● ● ● ● ●
●
● ●● ● ● ● ● ● ● ● ● ● ● ●●
●
● ● ●
●
● ●● ●
● ●
●
●
●
● ●● ●● ●
●
● ● ●
●
● ● ●●● ● ●
●
●
●
●
● ●
●●● ●
● ●
●
●
●
●
●
●
●
●
●
● ●
●
●
●
●
●
●
●●
●
●
● ●
● ●
●
●
●
● ●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
● ●
●
● ●
●
●
●
● ●
●
● ●
●
●
● ● ●
●
●
●
● ●
●
● ● ● ● ●
●
● ● ●
● ● ●
●● ●
●
0
2
4
6
8
10
−log10(p)
Clear Cell
● ●● ● ●● ● ● ● ● ● ● ● ●
●
●
● ● ● ● ●
●
● ●
●
● ● ●
●
● ● ●
●● ●
● ●
●
●
●
● ● ● ● ●
●●
●
● ● ●
●
●
●
●
●
●
●
●
●
● ●● ● ● ●
●
●
●●
●
● ● ●
● ● ● ●
●●
●
● ●
●
●●●●
●
●
● ●●
●
●
● ●
●
●
●
● ● ●●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
● ●
● ●
●
●
●● ●
●
●●
● ●●
● ● ● ● ● ●
●●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
●● ● ● ● ● ●● ●● ● ●● ● ● ● ● ● ●
●
●
● ● ● ● ●
●
●
●
● ● ● ● ● ● ●
●
● ●
● ● ● ●
●
●
●
● ● ● ● ●
●
●
●
●●
●
●
●
●
● ●
● ● ● ●●●● ●
●
● ● ●
●● ● ●
●
●●
●
●
●
●
● ●●
●
●
●
● ● ●●● ●
●
● ● ●
●
●
●●
●
● ● ● ●
●
●
●
● ●
●
● ● ● ● ● ● ● ●
●
●
●
●
● ●
●
● ●
●
● ● ●
●
● ●
●
● ●
● ●
● ●● ●
●
● ●
● ● ●● ●
●
●
●
●
● ● ●
● ● ●
●
● ●
●
●
● ●
● ●
●
●
●
●
●
●
●
● ●●
●
●
●
●
● ●
●
●
● ●
● ● ● ●
●●
● ●
● ● ●
●
● ● ●
●
●
● ●
●
● ●
● ●● ● ● ● ●
● ●
●
● ●●● ● ●● ● ● ●
●
● ●●
● ●
● ●
●
●
● ●
●
●
● ● ●
●
●
●
●
●
●
●
●
●●
●
●
● ●
●
●
●
●
● ●
●●
●
●
●
●
● ●
●
●●
●● ● ●● ●● ●●
● ● ● ●● ● ●
●
● ● ● ● ●
●
●
●
●
●
● ●
●
●
●
● ●
●
● ● ● ●
● ● ●
●
●
●
●
HNF1B
!
Figure 5-2: Genetic variants in the HNF1B locus are associated with risk of ovarian
cancer histological subtypes. Plotted in each panel is the –log10(P-value) from the SNP
association with risk for each subtype (Manhattan plots) located in the 150kb region described
in the text. Imputed SNPs are indicated with a relatively lighter color, while the genotyped SNPs
are indicated with a darker color. Dashed lines indicate the genome-wide significance threshold
(5x10
-8
). The LD plot on the bottom shows the r
2
between the SNPs. Genomic coordinates are
based on hg19 (Build37).
!
132
!
endometrioid subtypes (Figure 5-2).
Table 5-1: Association between invasive, serous and clear cell ovarian cancer for ten
HNF1B SNPs that reached genome-wide significance in Whites.
Minor alleles at nine SNPs, six genotyped and three imputed, were associated
with increased risk of invasive serous ovarian cancer at p<5x10
-8
(Table 5-1). The risk
signal spanned a 21·4kb region from the 5′ UTR through part of intron 4 of HNF1B
(Figure 5-2). The most strongly associated SNP for invasive serous ovarian cancer
(rs7405776, MAF 36%) conferred a 13% increased risk per minor allele (p=3·1 x 10
-10
,
Table 5-1). The signals of this SNP and the eight other genome-wide significant SNPs
were indistinguishable given the linkage disequilibrium and resulting haplotype structure.
For the clear cell subtype, rs11651755 (MAF 45%) was associated with a 23%
decreased risk of disease at a genome-wide significant level (p=2 x 10
-8
, Table 5-1). This
133
!
signal was distinct from the nine significant SNPs for invasive serous cancer (Table 5-1).
The odds against the serous-associated SNP, rs7405776, as the true best hit for clear
cell ovarian cancer were 244:1. Conversely the odds against the clear cell SNP,
rs11651755, as the true best hit for serous were 1808:1. Further, when rs11651755 and
Clear Cell
(n=17)
Serous
(n=196)
HNF1B IHC
A) HNF1B
CpGs
B) 1,003 CpG Loci
Across the Genome
Normal (n=7)
0%
1-50%
>50%
0 1
DNA Methylation
Beta Value
HNF1B IHC
CIMP
CIMP STATUS
No
Yes
Figure 5-3: HNF1B promoter DNA methylation, protein expression and global DNA
methylation pattern by subtype. Each row is a tissue sample collected at the Mayo Clinic that
belongs to one of the three categories: normal ovarian tissue (n=7), clear cell ovarian tumors
(n=17) or serous ovarian tumors (n=196); endometroid (n=49) and mucinous (n=7) tumors are
not included in this Figure. Each column represents a a CpG locus, either from the region
flanking the HNF1B transcription start site (Panel A, ordered by genomic locations with an
arrow indicating the transcription start site), or from a global panel of 1,003 CpG loci mapped to
autosomal CpG island regions that distinguish clear cell and serous subtypes (Panel B, ordered
by average DNA methylation across the samples). For each horizontal panel group, the
samples (rows) are ordered by HNF1B IHC status. The heatmap shows the DNA methylation
beta value, with blue indicating low DNA methylation and red indicating high methylation. Clear
cell tumors showed less DNA methylation at the HNF1B promoter region and correspondingly
higher HNF1B protein expression. The clear cell tumors generally show a CpG Island
Methylator Phenotype (CIMP) where there is extensive gain of aberrant promoter methylation
in a correlated manner. CIMP status (left side bar, defined as methylated at >80% of the 1,003
loci), and is highly correlated HNF1B expression. Also noteworthy is that the HNF1B promoter
DNA methylation (Panel A) is the opposite from the global pattern (Panel B, Supplementary
Fig. S8). This suggests HNF1B DNA methylation is not a passenger event of global DNA
methylation changes.
!
134
!
rs7405776 were jointly modeled, the signal for clear cell cancer was driven completely
by rs11651755, while that for the serous disease was driven by rs7405776 (Table 5-1).
The clear cell SNP (rs11651755) sits on five haplotypes, only three of which also contain
the serous SNP (rs7405776). Thus, different SNPs in the HNF1B gene regions explain
the associations observed for serous and clear cell ovarian cancer.
DNA Methylation and Protein Expression
*
*
*
*
*
*
*
*
*
*
*
*
*
*
−0.4 −0.2 0.0 0.2 0.4 0.6
0 2 4 6 8 10 12 14
104,033 CpG
Difference in Beta Value
BH−adjusted P value
* HNF1B Probes
Figure 5-4: HNF1B promoter methylation is unlikely to be a passenger event by global
DNA methylation changes. We compared the DNA methylation level at 104,033 CpG loci
that are unmethylated (beta value <0.2) in the 10 normal samples, in 254 serous tumors to
17 clear cell (Mayo panel) with two-sample t-test. The raw p values are adjusted with the
Benjamini-Hochberg method and the –log10 adjusted p values are plotted as the y axis in
the volcano plot, against mean beta value for clear cell minus mean beta value for serous,
as the x axis. The non-shaded area indicate adjusted p<0·05, absolute difference in beta
value > 0.2. A subset of 1,003 is used for Figure 3, with an even more stringent cut-off of the
adjusted p value <0.005, indicated by the dashed line. The red stars indicate the HNF1B
loci. We can see that while clear cell tumors generally have far more hypermethylation,
HNF1B is one of the few genes hypermethylated in the serous subtype. This argues against
the possibility that HNF1B hypermethylation in the serous subtype is a passenger event with
global hypermethylation.
!
135
!
The identification of HNF1B as a susceptibility gene for serous and clear cell
ovarian cancer led us to further evaluate the relationship between HNF1B promoter DNA
methylation, protein expression and histologic subtype. Immunohistochemistry (IHC)
analysis for HNF1B protein expression in 1,149 ovarian cancers from the Ovarian Tumor
Tissue Analysis Consortium (OTTA), and DNA methylation analysis on 269 of these
tumors revealed that the majority of clear cell tumors expressed the HNF1B protein, and
were unmethylated at the HNF1B promoter, while the majority of serous tumors lacked
HNF1B protein expression and displayed frequent HNF1B promoter methylation (Figure
5-3, Figure 5-4).
Although most clear cell tumors were devoid of HNF1B promoter methylation,
they revealed a surprisingly high frequency of CpG island hypermethylation at other sites
across the genome, indicative of a CpG Island Methylator Phenotype (CIMP). The few
clear cell tumors lacking HNF1B expression exhibited HNF1B promoter methylation, and
a correspondingly low frequency of CpG island methylation throughout the genome,
similar to the serous subtype (Figure 5-3). HNF1B expression and CIMP methylation are
strongly associated (Figure 5-3, p=3 x 10
-16
). Further, minimal hypermethylation is
observed in serous tumors overall, but HNF1B is one notable exception (Figure 5-4).
DNA Methylation and Genotype
136
!
We further investigated the relationship between risk allele genotypes and
HNF1B DNA methylation in 231 serous ovarian cancers. The top serous risk SNP,
a.
b.
c.
0%
100%
Figure 5-5: Correlation of serous risk-associated SNPs with HNF1B promoter DNA
methylation level. Plotted is the LD region defined as r
2
>0·2 with the top serous SNP
rs7405776. (a) Annotation of the region in terms of (from top to bottom:) UCSC genes,
FANTOM mark, Polycomb Repressive Complex marks (PRC2 and PRC1) (Ku et al.,
2008), the chromatin status determined in stem cells (Ernst et al., 2011), the conservation
score across this region, and the CpG island information, on top of the location of the
HM450 probe used in Panel B. (b) Boxplots of promoter DNA methylation level of HNF1B
(cg14487292) by SNP genotype with position indicated in Panel C. This DNA methylation
probe was selected based on inverse association with mRNA expression for HNF1B, and
does not contain any SNP with MAF>1% in its probe sequence. Each boxplot shows the
distribution of DNA methylation level by genotype (homozygous major – white;
heterozygous – gray; homozygous minor – black, where the minor alleles are the risk
alleles). Two-sided p values testing for trend are presented, and are computed for 231
Mayo Clinic high-grade, high-stage serous tumors to avoid confounding by histological
subtypes, and also to be consistent with the TCGA data (primarily high-grade, high-stage
serous). Results were similar with all subtypes combined. The risk alleles are associated
with significantly increased DNA methylation.
!
137
!
rs7405776, showed only a borderline association with increased promoter methylation
(p=0.07, Figure 5-5). Intriguingly, the association between SNPs in HNF1B and HNF1B
promoter DNA methylation strengthened as their location approached the promoter
region, and the strongest signal came from a few SNPs, exemplified by rs11658063,
overlapping with a Polycomb Repressive Complex 2 (PRC2) mark in embryonic stem
cells (p=0.003; Figure 5-5). We validated this SNP-methylation association in the TCGA
data (Figure 5-6). This association is across the entire promoter region (Figure 5-7).
None of the probes used contained common SNPs in the sequence, excluding technical
artifact as a confounder of this association.
rs3744763
p= 0.17
17−36092841
p= 0.1
rs7405776
p= 0.069
rs757210
p= 0.054
rs4239217
p= 0.0077
rs61612821
p= 0.32
rs11657964
p= 0.0031
rs7501939
p= 0.0031
rs11658063
p= 0.0026
rs3744763
p= 0.08
rs757210
p= 0.01
rs4239217
p= 0.08
rs7501939
p= 0.02
TCGA n=519
Mayo n=231
!
Figure 5-6: Validation of the SNP-DNA methylation association with TCGA data. Only
four out of the nine serous SNPs were available on the Illumina Human1M-Duo BeadChip
used in TCGA. The DNA methylation probe cg14487292 was not available on the
HumanMethylation27k platform, so cg02335804, located in the same promoter region, was
used as a surrogate. The color of each box indicate the genotypes, i.e., homozygous major
(white), heterozygous (gray) and homozygous minor (black), where the minor alleles are the
risk alleles. The p values for the Mayo data are two-sided trend p values and one-sided trend
p values for the validation.!
138
!
!
!
C*/C*
C*/G
G/G
a.
b.
!
Figure 5-7: HNF1B DNA methylation levels across the entire promoter region differ by
rs11658063 genotype. We further examined the DNA methylation level across the HNF1B
gene promoter region for different genotypes at rs11658063 (relative position indicated with a
red arrow; Mayo panel). (A) Similar to Figure 5-5, with the bottom panel showing the probe
locations for HumanMethylation450 and HumanMethylation27 platform. (B) A blow up of the
region flanking the transcription start site that is unmethylated in the normal tissue samples,
with two CpG islands associated. Shown is a heatmap where blue indicate low methylation
(Beta value=0) and red indicate high methylation (Beta value=1) for each of the loci
interrogated by HM450 at this region. The heatmap is subdivided into six subpanels, to
separate samples (rows) with the three different genotypes (star indicates the risk allele) at
rs11658063 (position indicated with a red arrow), and CpG loci (columns) as upstream and
downstream of the transcription start site. We can see that the genotypes at rs11658063
(location indicated with an red arrow) influences overall DNA methylation level, but the
influences are more pronounced for the upstream promoter region.
!
139
!
Overexpression of HNF1B
Given the proposed role of HNF1B in clear cell tumorigenesis, we stably
overexpressed the gene in immortalized endometriosis epithelial cells (EECs), which are
hypothesized to be a cell of origin for clear cell ovarian cancers (Pearce et al., 2012).
EECs overexpressing HNF1B acquired an enlarged, flattened morphology and multi-
nucleated cells accumulated in the cultures (Figure 5-8A). Also, significant upregulation
a.
b.
200 µm
Figure 5-8: Phenotypic effects and downstream targets of HNF1B overexpression in
immortalized endometriosis cells (EECs). (A) Morphological changes in EECs expressing a
HNF1B GFP fusion protein (EEC
GFP.HNF1B
). GFP positive cells were sorted using flow
cytometry. The arrows indicate five nuclei contained within a single EEC
GFP.HNF1B
cell, showing
the aberrant polynucleation we observed in these cells. Using flow cytometry we quantified the
increase in polynucleation in EEC
GFP.HNF1B
to be around 8-fold compared to controls (data not
shown). (B) Gene expression analysis of HNF1B target genes and clear cell ovarian cancer
associated genes. (* P>0.01).!
140
!
of HNF1B-associated genes SPP1, DPP4, and ACE2 was observed upon HNF1B
overexpression in EECs (Figure 5-8B).
141
!
Discussion
HNF1B appears to play a prominent role in ovarian cancer etiology. It is the first
clear cell ovarian cancer susceptibility gene identified and variation in the gene is also
associated with risk of serous ovarian cancer at a genome-wide significance level. The
gene is overexpressed in clear cell tumors and silenced in serous tumors. The strong
association between HNF1B expression and CIMP methylation (p=3 x 10
-16
), and the
reciprocal nature of DNA methylation at the HNF1B promoter CpG islands, versus other
CpG islands across the genome, suggests that HNF1B promoter methylation is not
merely a CIMP passenger event; in fact, HNF1B expression may even contribute to the
hypermethylation phenotype. Taken together, these data suggest differing roles for
HNF1B in these invasive EOC subtypes: a potential gain-of-function in clear cell ovarian
cancer and loss-of-function in serous ovarian cancer, underscoring the heterogeneity of
this disease.
Different SNPs in the HNF1B gene regions explain the associations observed for
serous and clear cell ovarian cancers. These different effects provide further support for
the growing view that the histological subtypes of ovarian cancer represent distinct
diseases (Crum et al., 2007; Gilks, 2010; Gounaris et al., 2011; Kobel et al., 2008;
Kurman and Shih Ie, 2010; Pearce et al., 2012; Risch et al., 1996) with endometriosis as
a proposed cell of origin for clear cell disease (Pearce et al., 2012) and fallopian tube
fimbriae as one for serous disease (Crum et al., 2007). Interestingly, no association was
observed between HNF1B genotypes and endometrioid ovarian cancer despite the view
that, like clear cell, endometriosis is also a cell of origin for this subtype. The lack of
association may be due to a different transformation mechanism from endometriosis for
the endometrioid subtype, given that although the HNF1B promoter remains
142
!
unmethylated in the endometrioid subtype, the endometrioid subtype does not over
express HNF1B. Alternatively, misclassification of high-grade serous EOC as high-grade
endometrioid could result in a bias toward the null for the endometrioid subtype.
Variation in the 5′ UTR through the intron 4 region of HNF1B is also associated
with susceptibility to prostate (Berndt et al., 2011; Gudmundsson et al., 2007; Sun et al.,
2008; Thomas et al., 2008) and uterine cancer (Spurdle et al., 2011) (where minor
Top Serous Ovarian Cancer
SNP
Clear Cell Ovarian Cancer SNP
Prostate and Endometrial
Cancer and Diabetes SNP
Diabetes SNP
Diabetes SNP
Prostate SNP
* * * * * * * * *
*Genome-wide significantly associated Serous SNPs
Figure 5-9: Linkage disequilibrium (LD) plot of the genome-wide significant serous*
(n=9) and clear cell (n=1) SNPs as well as the SNPs associated with prostate and
uterine cancer and diabetes. The r2 value between the SNPs is given in each box based on
the 1000 Genome Project. The highest r2 between the serous SNPs and the clear cell SNP,
rs11651755, is 0.70. The r2 values between the top serous SNP, rs7405776, and the other
eight range from 0.24 to 0.97.
!
143
!
alleles of certain SNPs are associated with decreased risk) and type II diabetes
(Gudmundsson et al., 2007; Winckler et al., 2007) (increased risk for the same or
correlated SNP alleles; Figure 5-9). The opposing directions of these associations mirror
the differential effects seen here in ovarian cancer. The most strongly associated SNP
for both prostate (Berndt et al., 2011) and uterine cancer (Spurdle et al., 2011) is
rs4430796, correlated at r
2
=0.94 with the top clear cell ovarian cancer SNP, rs11651755,
suggesting a common risk variant. Although increased risk of Type II diabetes has been
reported with rs4430796 (Gudmundsson et al., 2007), Winckler and colleagues
(Winckler et al., 2007) have suggested that the best marker of diabetes risk is rs757210,
which correlated with an r
2
of 0.97 with our top serous SNP. Thus, the evidence
suggests that a specific variant(s) in HNF1B predisposes to clear cell ovarian, uterine,
and prostate cancers and that a different variant(s) is associated with diabetes and
serous ovarian cancer.
We were able to completely fine-map the HNF1B region, localize the signal and
identify a handful of potentially causal SNPs. This is quite different from other regions of
the genome where it is not uncommon to identify hundreds of candidate causal SNPs.
Further, an important link often missing when susceptibility loci are identified is the
functional role the variant plays in disease. In the case of serous ovarian cancer, the
SNP-HNF1B promoter DNA methylation association strengthens as it approaches the
promoter region, particularly where it overlaps with a PRC2 mark. PRC2–DNA
methyltransferase (DNMT) crosstalk has been proposed to be a mechanism of
predisposition to cancer-specific hypermethylation (Widschwendter et al., 2007). Our
DNA methylation data suggest that the causal risk allele(s) for the serous subtype may
predispose the promoter to acquiring aberrant methylation, thereby promoting the
144
!
development of serous but not clear cell tumors. This predisposition could be a direct
functional effect of the SNP on the DNA methylation machinery, or could act indirectly
through differential binding affinity for PRC2 or one or more transcription factors. Given
that we were able to fine-map the HNF1B region, it is unlikely that an unidentified
common variant explains these associations. For serous ovarian cancer, the methylation
signal suggests that the causal variant is most likely to be among those located within
the region with the PRC2 mark for which we identified five SNPs with genome-wide
significance.
This is the first study investigating the effects of overexpression of HNF1B in
endometriosis, and the results support the hypothesis that HNF1B may have an
oncogenic role in the initiation of clear cell ovarian cancers, as speculated by Gounaris
and colleagues as a key step of endometriosis transformation (Gounaris et al., 2011).
The observation in our data that HNF1B induces a polynucleated phenotype in EEC
cells is intriguing, as clear cell ovarian cancers are often tetradiploid, more so than other
ovarian cancer subtypes (Skirnisdottir et al., 2005). The polynucleated phenotype may
suggest that HNF1B overexpression in EECs perturbs cytokinesis, causing aneuploidy in
some cells.
Histology re-review of the three clear cell tumors that do not express HNF1B
revealled two scenarios: two samples with inconsistent evaluations between pathologists
and one consistently called clear cell. They might be cases that are especially difficult to
classify, and therefore a molecular signature, e.g., CIMP or HNF1B status, would be of
great help in correctly classifying those tumors. The one sample that is called
consistently clear cell tumor but does not express HNF1B might represent a rare
145
!
subtype of clear cell carcinoma. With a larger cohort of clear cell ovarian cancers these
possibilities can be investigated.
To our knowledge, this is the first report of tumor DNA methylation patterns
leading to the identification of a germline susceptibility locus, underscoring the value of
TCGA. Recent studies suggest a strong genetic component to inter-individual variation
in tumor DNA methylation, and demonstrate both cis- and trans- associations between
genotypes and DNA methylation (Bell et al., 2011). In addition, methylation quantitative
trait loci (mQTLs) were found to be enriched for expression QTLs (eQTLs). It has also
been shown that epimutation is associated with genetic variation, for example
associations have been demonstrated between 5’UTR MLH1 variants and MLH1
epigenetic silencing (Hitchins et al., 2011). Moreover, we have for the first time
demonstrated the existence of a CIMP phenotype in ovarian cancer, highlighting the
complicated nature of the disease.
In summary, variation in HNF1B is associated with serous and clear cell
subtypes of ovarian cancer in opposite manner at genetic, epigenetic, and protein
expression levels. These observations are compatible with a tumor suppressor role in
serous cancer and an oncogenic role in clear cell disease. Future efforts should focus on
understanding these mechanisms as they could have major clinical implications for
ovarian cancer, based on better subtype stratification, potential novel treatment
approaches, and a better understanding of disease etiology. Currently effective
chemotherapeutics for clear cell ovarian cancer is lacking, but our study reveals that
HNF1B-expressing clear cell tumors have extensive epigenetic alterations that
potentially make them good candidate for epigenetic therapies.
146
!
Methods
Molecular Aspects
TCGA Data Access
We downloaded the TCGA serous ovarian cancer data packages from the TCGA
public-access ftp:
https://tcga-data.nci.nih.gov/tcgafiles/ftp_auth/distro_ftpusers/anonymous/tumor/ov/.
Data generated with the following platforms were used: Affymetrix HT Human Genome
U133 Array Plate Set; Agilent 244K Custom Gene Expression G4502A-07-3; Affymetrix
Human Exon 1.0 ST Array; and Illumina Infinium HumanMethylation27 Beadchip. The
Illumina Human1M-Duo DNA Analysis BeadChip Genotype data were downloaded from
the controlled access data tier. A full list of the TCGA packages used in this study:
Affymetrix HT Human Genome U133 Array Plate Set
broad.mit.edu_OV.HT_HG-U133A.Level_3.11.1007.0/
broad.mit.edu_OV.HT_HG-U133A.Level_3.12.1007.0/
broad.mit.edu_OV.HT_HG-U133A.Level_3.13.1007.0/
broad.mit.edu_OV.HT_HG-U133A.Level_3.14.1007.0/
broad.mit.edu_OV.HT_HG-U133A.Level_3.15.1007.0/
broad.mit.edu_OV.HT_HG-U133A.Level_3.17.1007.0/
broad.mit.edu_OV.HT_HG-U133A.Level_3.18.1007.0/
broad.mit.edu_OV.HT_HG-U133A.Level_3.19.1007.0/
broad.mit.edu_OV.HT_HG-U133A.Level_3.21.1007.0/
broad.mit.edu_OV.HT_HG-U133A.Level_3.22.1007.0/
broad.mit.edu_OV.HT_HG-U133A.Level_3.24.1007.0/
147
!
broad.mit.edu_OV.HT_HG-U133A.Level_3.27.1007.0/
broad.mit.edu_OV.HT_HG-U133A.Level_3.40.1007.0/
broad.mit.edu_OV.HT_HG-U133A.Level_3.9.1007.0/
Agilent 244K Custom Gene Expression G4502A-07-3
unc.edu_OV.AgilentG4502A_07_3.Level_3.1.5.0/
unc.edu_OV.AgilentG4502A_07_3.Level_3.2.0.0/
Affymetrix Human Exon 1.0 ST Array
lbl.gov_OV.HuEx-1_0-st-v2.Level_3.11.2.0/
lbl.gov_OV.HuEx-1_0-st-v2.Level_3.12.2.0/
lbl.gov_OV.HuEx-1_0-st-v2.Level_3.13.2.0/
lbl.gov_OV.HuEx-1_0-st-v2.Level_3.14.2.0/
lbl.gov_OV.HuEx-1_0-st-v2.Level_3.15.2.0/
lbl.gov_OV.HuEx-1_0-st-v2.Level_3.17.2.0/
lbl.gov_OV.HuEx-1_0-st-v2.Level_3.18.2.0/
lbl.gov_OV.HuEx-1_0-st-v2.Level_3.19.2.0/
lbl.gov_OV.HuEx-1_0-st-v2.Level_3.21.2.0/
lbl.gov_OV.HuEx-1_0-st-v2.Level_3.22.1.0/
lbl.gov_OV.HuEx-1_0-st-v2.Level_3.27.0.0/
lbl.gov_OV.HuEx-1_0-st-v2.Level_3.40.1.0/
lbl.gov_OV.HuEx-1_0-st-v2.Level_3.9.2.0/
Illumina Infinium HumanMethylation27 Beadchip
jhu-usc.edu_OV.HumanMethylation27.Level_3.1.4.0/
jhu-usc.edu_OV.HumanMethylation27.Level_3.10.0.0/
jhu-usc.edu_OV.HumanMethylation27.Level_3.11.1.0/
148
!
jhu-usc.edu_OV.HumanMethylation27.Level_3.12.1.0/
jhu-usc.edu_OV.HumanMethylation27.Level_3.13.0.0/
jhu-usc.edu_OV.HumanMethylation27.Level_3.2.4.0/
jhu-usc.edu_OV.HumanMethylation27.Level_3.3.4.0/
jhu-usc.edu_OV.HumanMethylation27.Level_3.4.3.0/
jhu-usc.edu_OV.HumanMethylation27.Level_3.5.2.0/
jhu-usc.edu_OV.HumanMethylation27.Level_3.6.2.0/
jhu-usc.edu_OV.HumanMethylation27.Level_3.7.2.0/
jhu-usc.edu_OV.HumanMethylation27.Level_3.8.2.0/
jhu-usc.edu_OV.HumanMethylation27.Level_3.9.0.0/
DNA Methylation Data Production For the OCRF Tumor Panel
The Illumina Infinium HumanMethylation27 assay was performed as
described(TCGA, 2011)!on 32 serous and four clear cell ovarian tumors from USC Norris
Comprehensive Cancer Center and Duke University (‘OCRF Tumor Panel’). The beta
values for each sample and locus were calculated with mean non-background corrected
methylated (M) and unmethylated (U) signal intensities with the formula M/(M+U),
representing the percentage of methylated alleles. Detection p-values were calculated
by comparing the set of analytical probe replicates for each locus to the set of 16
negative control probes. Data points with detection p-values > 0.05 were masked.
DNA Methylation Data Production for the Mayo Tumor Panel
We also performed the Infinium HumanMethylation450 BeadChip assay on an
independent set of tumor DNA in the Mayo Clinic Genotyping Shared Facility using
recommended Illumina protocol (Bibikova et al., 2011). 1µg of tumor DNA was bisulfite-
converted using the Zymo EZ96 DNA Methylation Kit. Three samples failing quality
149
!
control were removed, leaving DNA methylation data on 333 ovarian cancer cases,
including 254 serous and 17 clear cell tumors. Plate normalization was done with a
linear model on the logit-transformed beta values, following back-transformation to the
(0,1) range. The quality of the bisulfite converted DNA and the performance of the CpG
probes were assessed using a CEPH control, a whole genome amplified (WGA)
negative control and placental positive control samples. Internal placental positive
control, WGA negative control, and a CEPH control were used for quality control and the
mean intra-class correlation across the two batches of samples was 0.90, 0.96 and 0.99,
respectively. Intra-class correlation for ovarian duplicate samples was > 0.99.
Immunohistochemistry Assay
Previously built tissue microarrays (TMA), triplicate core, 0·6 mm were cut at
4µm thickness and mounted on superfrost slides. Slides were stained on a Ventana
Benchmark XT using the manufacturer’s pretreatment protocol CC1-standard. A
pathologist (MK) evaluated the immunohistochemistry staining, and assign the sample a
score 0 in the absence of any nuclear staining, score 1 for any nuclear staining >1% to
50%, or score 2 for more than 50% tumor cell nuclei positive for HNF1B.
Genotype and DNA Methylation Association
We assessed the correlation of germline genotype at the nine genome-wide
significant SNPs in serous cancer, with HNF1B DNA promoter methylation status using
the Mayo Tumor Panel. Probe cg14487292 was used as it was most inversely correlated
with mRNA expression. The nominal p-values are from two-sided tests for linear trend in
the DNA methylation beta values across the three genotypes for each locus. Bonferroni
adjustment was not done for multiple comparisons as the SNPs are highly correlated.
Validation was done with the TCGA data with 519 tumors. Four out of nine SNPs are
150
!
available on the TCGA platform. For promoter DNA methylation, cg02335804 was used
as a surrogate since cg14487292 (the two CpGs are 278bp away) was not present on
the HumanMethylation27 platform. The p-values are from one-sided tests for linear trend
in the DNA methylation beta value across the three genotypes for each locus. The
nominal Bonferroni adjusted p-value cutoff would be 0.013 (0.05/4).
In vitro model of HNF1B overexpression
An immortalized endometriosis epithelial cell (EEC) line was generated by
lentiviral transduction of hTERT (Addgene plasmid 12245) into primary EECs. TERT
immortalized EECs were transduced with lentiviral HNF1B-GFP or GFP (Genecopeia)
supernatants and positive cells selected with 400ng/ml puromycin (Sigma). GFP
expression was confirmed by fluorescent microscopy; HNF1B expression was confirmed
by real-time PCR.
For gene expression studies, RNA was harvested from cells using the QIAgen
RNeasy kit with on-column DNase I digestion. 1µg RNA was reverse transcribed using
an MMLV reverse transcriptase enzyme (Promega), and relative mRNA level assayed
using the ABI 7900HT FAST Real-Time PCR system utilizing the delta-delta Ct method.
Statistical analyses were performed using Prism. Two-tailed paired t-tests with
significance level of 0·05 were used.
Genomic/Epigenomic Data Analysis and Visualization
The statistical analyses were done in R (version 2.15.0). Mapping and
characterization of the HumanMethylation450 probes were done with the R package
IlluminaHumanMethylation450k.db. The UCSC tracks were downloaded with rtracklayer
(Lawrence et al., 2009) . The PRC1 (Ring1b) and PRC2 mark (H3K27me3) ChIP-seq
and the chromatin state data (ChromHMM) were from previous work in embryonic stem
151
!
cells. The genomic and epigenomic data were mapped to the genome with
Build37(hg19) coordinates, and visualized using the R packages with GenomicRange
(Aboyoun et al., R package) and Gviz (Hahne et al., R package).
Genetic Association Study
Study Design
The genetic susceptibility aspect of this study was organized by the Collaborative
Oncological Gene-Environment Study (COGS), an ovarian, breast and prostate cancer
consortium. The ovarian cancer part of this effort on which the current report is based is
led by the Ovarian Cancer Association Consortium (OCAC) and included 43 studies
(Appendix B). Following sample quality control, 44,308 subjects including 16,111
patients with invasive EOC, 2063 with low malignant potential (borderline) disease and
26,134 controls were available for analysis; results presented here are restricted to
invasive cancers. All studies obtained approval from their respective human research
ethics committees, and all participants provided written informed consent.
Selection of SNPs
Tagging SNPs were selected in the HNF1B region using the program SNAGGER
(Edlund et al., 2008) from the International HapMap Project CEU population (White) in
order to cover all SNPs in the region with a minor allele frequency of 0.05 with an r
2
of
0.80. This resulted in the selection of 40 SNPs. In addition, because of the association
between prostate cancer and HNF1B, an additional 134 SNPs were selected by
The Prostate Cancer Association Group to Investigate Cancer Associated Alterations in
the Genome (The PRACTICAL Consortium (Eeles et al., 2009; Kote-Jarai et al., 2008;
Kote-Jarai et al., 2011)) to provide full fine-mapping information based on 174 genotyped
SNPs. A 150kb-region surrounding HNF1B was identified for fine-mapping (hg18
152
!
coordinates 33,100,000-33,250,000). Fine-mapping SNPs were selected at this locus
from the March 2010 (Build 36) release of the 1000 Genomes Project for all known
SNPs with minor allele frequency >0.02 in Europeans and r
2
>0.1 with the reported
prostate cancer associated SNPs (s11649743 and rs4430796). In addition, phase I
haplotype data from the 1000 Genomes Project (January 2012) were used to impute
genotypes for SNPs across this region, resulting in available data on an additional 307
SNPs with minor allele frequency >0.02 in European Whites and imputation r
2
>0.30
(IMPUTE 2.2).
SNP Genotyping
OCAC genotyping was conducted by McGill University and Génome Québec
Innovation Centre (n=19,806) and the Mayo Clinic Medical Genome Facility (n=27,824)
using an Illumina Infinium iSelect BeadChip. Genotypes were called using GenCall.
Each 96-well plate contained 250 ng genomic DNA (or 500 ng whole-genome amplified
DNA). Raw intensity data files for all consortia were sent to the COGS data co-
ordination centre at the University of Cambridge for centralized genotype calling and QC.
Initial calling used a cluster file generated using 270 samples from
Hapmap2. These calls were used for ongoing QC checks during the genotyping. To
generate the final calls used for the data analysis, we first selected a subset of 3,018
individuals, including samples from each of the genotyping centers, each of the
participating consortia, and each major ethnicity. Only plates with a consistent high call
rate in the initial calling were used. The HapMap samples and ~160 samples that were
known positive controls for rare variants on the array were used to generate a cluster file
that was then applied to call the genotypes for the remaining samples. We also
investigated two other calling algorithms: Illumnus (Teo et al., 2007) and GenoSNP
153
!
(Giannoulatou et al., 2008), but manual inspection of a sample of SNPs with discrepant
calls indicated that GenCall was invariably superior.
Sample QC for Genotyping
1,273 OCAC samples were genotyped in duplicate. Genotypes were discordant
for greater than 40 percent of SNPs for 22 pairs. For the remaining 1,251 pairs,
concordance was greater than 99.6 percent. In addition we identified 245 pairs of
samples that were unexpected genotypic duplicates. Of these, 137 were phenotypic
duplicates and judged to be from the same individual. We used identity-by-state to
identify 618 pairs of first-degree relatives. Samples were excluded according to the
following criteria: 1) 1,133 samples with a conversion rate of less than 95%; 2) 169
samples with heterozygosity >5 standard deviations from the intercontinental ancestry
specific mean heterozygosity; 3) 65 samples with ambiguous sex; 4) 269 samples with
the lowest call rate from a first-degree relative pair 5) 1,686 samples that were either
duplicate samples that were non-concordant for genotype or genotypic duplicates that
were not concordant for phenotype. Thus, a total of 44,308 subjects including 16,111
invasive cases, 2,063 borderline cases and 26,134 controls were available for analysis.
SNP Quality Control
In total, 211,155 SNP assays, identified across a number of studies, were
successfully designed and included on the array. SNPs were excluded according to the
following criteria: (1) 1,311 SNPs without a genotype call; (2) 2,857 monomorphic SNPs;
(3) 5,201 SNPs with a call rate less than 95 percent and MAF > 0·05 or call rate less
than 99 percent with MAF < 0.05; (4) 2,194 SNPs showing evidence of deviation of
genotype frequencies from Hardy-Weinberg equilibrium (P<10
-7
); (5) 22 SNPS with
154
!
greater than two percent discordance in duplicate pairs. Overall, 94·5 percent passed
QC. Genotype clusters were visually inspected for the most strongly associated SNPs.
Statistical analysis
Subjects with greater than 90 percent European ancestry were classified as
European (n=39,944) and those with greater than 80 percent Asian and African ancestry
were classified as being Asian (n=2,388) and African respectively (n=387). All other
subjects were classified as mixed ancestry (n=1,770). We then used a set of 37,000
additional genotyped markers not suspected to be related to ovarian cancer risk to
perform principal components analysis within each major population subgroup(Price et
al., 2006). To enable this analysis on very large-scale samples we used an in-house
program written in C++ using the Intel MKL libraries for eigenvectors (available at
http://ccge.medschl.cam.ac.uk/software/).
We used the program LAMP (Li et al., 2005) for principal components analysis
(PCA) to assign intercontinental ancestry based on the HapMap (release no. 22)
genotype frequency data for European, African and Asian populations. For LAMP-
derived European ancestry groups for all patients of invasive cancer and for those with
serous invasive cancer, we carried out unconditional logistic regression analyses within
each study site, adjusted for the first five eigenvalues from the PCA for European
ancestry and then used a fixed-effects meta-analytic approach to obtain the summary
odds ratio estimate, 95% confidence interval and p-value. Log-additive mode of
inheritance was modeled (i.e., co-dominant), treating each SNP as an ordinal variable.
For the non-European groups for all invasive cases and serous cases as well as
for all groups for the other subtypes, we were not able to carry out within study analyses
due to the small sample sizes available. We thus conducted unconditional logistic
155
!
regression models adjusted for the first five principal components for the European
ancestry and the first two principal components for the other ancestry groups as well as
study site.
To evaluate the independence of associations between the top serous and clear
cell SNPs, we fit separate models by histology that contained both SNPs. In addition,
two correlated SNPs were found to be associated with both serous and clear cell
subtypes of ovarian cancer, with one SNP being more strongly associated with serous
(rs7405776) and the other more strongly associated with clear cell (rs11651755). It is
conceivable that the associations for both sub-types are being driven by the same SNP,
but, by chance, the other correlated SNP is giving a stronger signal for one of the sub-
types. We therefore compared the log-likelihood statistics logistic regression models for
each SNP with each subtype. The odds in favor of one SNP being the driver of the
signal is given as exp(log-likelihood
SNP1
- log-likelihood
SNP2
).
For haplotype analysis, we used the tagSNPs program (Stram et al., 2003) to
obtain the haplotype dosage for each subject for the LAMP-derived European ancestry
group for haplotypes with a frequency of 1% or greater. The associations between
haplotype and risks of serous and clear cell ovarian cancer were modelled by meta-
analysis relative to the most common haplotype. The region for haplotype analysis was
defined as extending to the point around the top serous SNP, rs7405776, where there
were no SNPs with an r
2
>0.20 with a minor allele frequency of 5%.
156
!
Chapter Five Bibliography
Aboyoun, P., Pages, H., and Lawrence, M. (R package). GenomicRanges:
Representation and manipulation of genomic intervals, pp. Version 1.8.3.
Antoniou, A.C., and Easton, D.F. (2006). Risk prediction models for familial breast
cancer. Future Oncol 2, 257-274.
Auranen, A., Pukkala, E., Makinen, J., Sankila, R., Grenman, S., and Salmi, T. (1996).
Cancer incidence in the first-degree relatives of ovarian cancer patients. Br J Cancer 74,
280-284.
Bell, J.T., Pai, A.A., Pickrell, J.K., Gaffney, D.J., Pique-Regi, R., Degner, J.F., Gilad, Y.,
and Pritchard, J.K. (2011). DNA methylation patterns associate with genetic and gene
expression variation in HapMap cell lines. Genome Biol 12, R10.
Berndt, S.I., Sampson, J., Yeager, M., Jacobs, K.B., Wang, Z., Hutchinson, A., Chung,
C., Orr, N., Wacholder, S., Chatterjee, N., et al. (2011). Large-scale fine mapping of the
HNF1B locus and prostate cancer risk. Hum Mol Genet 20, 3322-3329.
Bibikova, M., Barnes, B., Tsan, C., Ho, V., Klotzle, B., Le, J.M., Delano, D., Zhang, L.,
Schroth, G.P., Gunderson, K.L., et al. (2011). High density DNA methylation array with
single CpG site resolution. Genomics 98, 288-295.
Bolton, K.L., Ganda, C., Berchuck, A., Pharaoh, P.D., and Gayther, S.A. (2012). Role of
common genetic variants in ovarian cancer susceptibility and outcome: progress to date
from the Ovarian Cancer Association Consortium (OCAC). Journal of internal medicine
271, 366-378.
Bolton, K.L., Tyrer, J., Song, H., Ramus, S.J., Notaridou, M., Jones, C., Sher, T., Gentry-
Maharaj, A., Wozniak, E., Tsai, Y.Y., et al. (2010). Common variants at 19p13 are
associated with susceptibility to ovarian cancer. Nat Genet 42, 880-884.
Crum, C.P., Drapkin, R., Miron, A., Ince, T.A., Muto, M., Kindelberger, D.W., and Lee, Y.
(2007). The distal fallopian tube: a new model for pelvic serous carcinogenesis. Curr
Opin Obstet Gynecol 19, 3-9.
157
!
Edlund, C.K., Lee, W.H., Li, D., Van Den Berg, D.J., and Conti, D.V. (2008). Snagger: a
user-friendly program for incorporating additional information for tagSNP selection. BMC
Bioinformatics 9, 174.
Eeles, R.A., Kote-Jarai, Z., Al Olama, A.A., Giles, G.G., Guy, M., Severi, G., Muir, K.,
Hopper, J.L., Henderson, B.E., Haiman, C.A., et al. (2009). Identification of seven new
prostate cancer susceptibility loci through a genome-wide association study. Nat Genet
41, 1116-1121.
Ernst, J., Kheradpour, P., Mikkelsen, T.S., Shoresh, N., Ward, L.D., Epstein, C.B.,
Zhang, X., Wang, L., Issner, R., Coyne, M., et al. (2011). Mapping and analysis of
chromatin state dynamics in nine human cell types. Nature 473, 43-49.
Giannoulatou, E., Yau, C., Colella, S., Ragoussis, J., and Holmes, C.C. (2008).
GenoSNP: a variational Bayes within-sample SNP genotyping algorithm that does not
require a reference population. Bioinformatics 24, 2209-2214.
Gilks, C.B. (2010). Molecular abnormalities in ovarian cancer subtypes other than high-
grade serous carcinoma. J Oncol 2010, 740968.
Goode, E.L., Chenevix-Trench, G., Song, H., Ramus, S.J., Notaridou, M., Lawrenson,
K., Widschwendter, M., Vierkant, R.A., Larson, M.C., Kjaer, S.K., et al. (2010). A
genome-wide association study identifies susceptibility loci for ovarian cancer at 2q31
and 8q24. Nat Genet 42, 874-879.
Gounaris, I., Charnock-Jones, D.S., and Brenton, J.D. (2011). Ovarian clear cell
carcinoma--bad endometriosis or bad endometrium? J Pathol 225, 157-160.
Gudmundsson, J., Sulem, P., Steinthorsdottir, V., Bergthorsson, J.T., Thorleifsson, G.,
Manolescu, A., Rafnar, T., Gudbjartsson, D., Agnarsson, B.A., Baker, A., et al. (2007).
Two variants on chromosome 17 confer prostate cancer risk, and the one in TCF2
protects against type 2 diabetes. Nat Genet 39, 977-983.
Hahne, F., Durinck, S., Ivanek, R., and Mueller, A. (R package). Gviz: Plotting data and
annotation information along genomic coordinates, pp. Version 1.0.0.
Hitchins, M.P., Rapkins, R.W., Kwok, C.T., Srivastava, S., Wong, J.J., Khachigian, L.M.,
Polly, P., Goldblatt, J., and Ward, R.L. (2011). Dominantly inherited constitutional
epigenetic silencing of MLH1 in a cancer-affected family is linked to a single nucleotide
variant within the 5'UTR. Cancer cell 20, 200-213.
158
!
Horikawa, Y., Iwasaki, N., Hara, M., Furuta, H., Hinokio, Y., Cockburn, B.N., Lindner, T.,
Yamagata, K., Ogata, M., Tomonaga, O., et al. (1997). Mutation in hepatocyte nuclear
factor-1 beta gene (TCF2) associated with MODY. Nat Genet 17, 384-385.
Kobel, M., Kalloger, S.E., Boyd, N., McKinney, S., Mehl, E., Palmer, C., Leung, S.,
Bowen, N.J., Ionescu, D.N., Rajput, A., et al. (2008). Ovarian carcinoma subtypes are
different diseases: implications for biomarker studies. PLoS medicine 5, e232.
Kote-Jarai, Z., Easton, D.F., Stanford, J.L., Ostrander, E.A., Schleutker, J., Ingles, S.A.,
Schaid, D., Thibodeau, S., Dork, T., Neal, D., et al. (2008). Multiple novel prostate
cancer predisposition loci confirmed by an international study: the PRACTICAL
Consortium. Cancer Epidemiol Biomarkers Prev 17, 2052-2061.
Kote-Jarai, Z., Olama, A.A., Giles, G.G., Severi, G., Schleutker, J., Weischer, M.,
Campa, D., Riboli, E., Key, T., Gronberg, H., et al. (2011). Seven prostate cancer
susceptibility loci identified by a multi-stage genome-wide association study. Nat Genet
43, 785-791.
Ku, M., Koche, R.P., Rheinbay, E., Mendenhall, E.M., Endoh, M., Mikkelsen, T.S.,
Presser, A., Nusbaum, C., Xie, X., Chi, A.S., et al. (2008). Genomewide analysis of
PRC1 and PRC2 occupancy identifies two classes of bivalent domains. PLoS genetics
4, e1000242.
Kurman, R.J., and Shih Ie, M. (2010). The origin and pathogenesis of epithelial ovarian
cancer: a proposed unifying theory. Am J Surg Pathol 34, 433-443.
Lawrence, M., Gentleman, R., and Carey, V. (2009). rtracklayer: an R package for
interfacing with genome browsers. Bioinformatics 25, 1841-1842.
Li, M., Boehnke, M., and Abecasis, G.R. (2005). Joint modeling of linkage and
association: identifying SNPs responsible for a linkage signal. Am J Hum Genet 76, 934-
949.
Lichtenstein, P., Holm, N.V., Verkasalo, P.K., Iliadou, A., Kaprio, J., Koskenvuo, M.,
Pukkala, E., Skytthe, A., and Hemminki, K. (2000). Environmental and heritable factors
in the causation of cancer--analyses of cohorts of twins from Sweden, Denmark, and
Finland. N Engl J Med 343, 78-85.
Pearce, C.L., Templeman, C., Rossing, M.A., Lee, A., Near, A.M., Webb, P.M., Nagle,
C.M., Doherty, J.A., Cushing-Haugen, K.L., Wicklund, K.G., et al. (2012). Association
159
!
between endometriosis and risk of histological subtypes of ovarian cancer: a pooled
analysis of case-control studies. Lancet Oncol 13, 385-394.
Pharaoh, P.D.P., Tsai, Y.-Y., Ramus, S.J., Phelan, C.M., Goode, E.L., Lawrenson, K.,
Price, M., Fridley, B.L., Tyrer, J.P., Shen, H., et al. (In Press). GWAS meta-analysis and
replication identifies three novel common susceptibility loci for ovarian cancer. Nature
Genetics.
Price, A.L., Patterson, N.J., Plenge, R.M., Weinblatt, M.E., Shadick, N.A., and Reich, D.
(2006). Principal components analysis corrects for stratification in genome-wide
association studies. Nat Genet 38, 904-909.
Risch, H.A., Marrett, L.D., Jain, M., and Howe, G.R. (1996). Differences in risk factors for
epithelial ovarian cancer by histologic type. Results of a case-control study. Am J
Epidemiol 144, 363-372.
Skirnisdottir, I., Seidal, T., Karlsson, M.G., and Sorbe, B. (2005). Clinical and biological
characteristics of clear cell carcinomas of the ovary in FIGO stages I-II. Int J Oncol 26,
177-183.
Song, H., Ramus, S.J., Tyrer, J., Bolton, K.L., Gentry-Maharaj, A., Wozniak, E., Anton-
Culver, H., Chang-Claude, J., Cramer, D.W., DiCioccio, R., et al. (2009). A genome-wide
association study identifies a new ovarian cancer susceptibility locus on 9p22.2. Nat
Genet 41, 996-1000.
Spurdle, A.B., Thompson, D.J., Ahmed, S., Ferguson, K., Healey, C.S., O'Mara, T.,
Walker, L.C., Montgomery, S.B., Dermitzakis, E.T., Australian National Endometrial
Cancer Study, G., et al. (2011). Genome-wide association study identifies a common
variant associated with risk of endometrial cancer. Nat Genet 43, 451-454.
Stram, D.O., Leigh Pearce, C., Bretsky, P., Freedman, M., Hirschhorn, J.N., Altshuler,
D., Kolonel, L.N., Henderson, B.E., and Thomas, D.C. (2003). Modeling and E-M
estimation of haplotype-specific relative risks from genotype data for a case-control
study of unrelated individuals. Human heredity 55, 179-190.
Sun, J., Zheng, S.L., Wiklund, F., Isaacs, S.D., Purcell, L.D., Gao, Z., Hsu, F.C., Kim,
S.T., Liu, W., Zhu, Y., et al. (2008). Evidence for two independent prostate cancer risk-
associated loci in the HNF1B gene at 17q12. Nat Genet 40, 1153-1155.
160
!
TCGA, C.G.A.R.N. (2011). Integrated genomic analyses of ovarian carcinoma. Nature
474, 609-615.
Teo, Y.Y., Inouye, M., Small, K.S., Gwilliam, R., Deloukas, P., Kwiatkowski, D.P., and
Clark, T.G. (2007). A genotype calling algorithm for the Illumina BeadArray platform.
Bioinformatics 23, 2741-2746.
Thomas, G., Jacobs, K.B., Yeager, M., Kraft, P., Wacholder, S., Orr, N., Yu, K.,
Chatterjee, N., Welch, R., Hutchinson, A., et al. (2008). Multiple loci identified in a
genome-wide association study of prostate cancer. Nat Genet 40, 310-315.
Tsuchiya, A., Sakamoto, M., Yasuda, J., Chuma, M., Ohta, T., Ohki, M., Yasugi, T.,
Taketani, Y., and Hirohashi, S. (2003). Expression profiling in ovarian clear cell
carcinoma: identification of hepatocyte nuclear factor-1 beta as a molecular marker and
a possible molecular target for therapy of ovarian clear cell carcinoma. Am J Pathol 163,
2503-2512.
Widschwendter, M., Fiegl, H., Egle, D., Mueller-Holzner, E., Spizzo, G., Marth, C.,
Weisenberger, D.J., Campan, M., Young, J., Jacobs, I., et al. (2007). Epigenetic stem
cell signature in cancer. Nat Genet 39, 157-158.
Winckler, W., Weedon, M.N., Graham, R.R., McCarroll, S.A., Purcell, S., Almgren, P.,
Tuomi, T., Gaudet, D., Bostrom, K.B., Walker, M., et al. (2007). Evaluation of common
variants in the six known maturity-onset diabetes of the young (MODY) genes for
association with type 2 diabetes. Diabetes 56, 685-693.
!
161
Conclusion
Molecular Subtyping of Cancer and Cross-Cancer-Type Similarity
The heterogeneity of many cancers is not fully explained by the current
classification by site of occurrence, morphology, stage and/or grade. Molecular
subtyping may aid in the diagnosis, treatment and prognosis of the disease. We found
distinct molecular subtypes of endometrial cancer, but high-stage, high-grade serous
ovarian cancer and clear cell renal cell carcinoma studied as part of the Cancer Genome
Atlas (TCGA) each seem to reflect one disease entity, with the heterogeneity in the clear
cell renal cell carcinoma largely explained by clinical stage and grade.
The fact that serous and serous-like endometrial tumors (DNA methylation
cluster 3 and basal-like breast tumors molecularly resemble high-grade serous ovarian
tumors, suggest that these three disease may be combatable with the same treatment
scheme.
Distinct molecular subtypes may also arise from different cells of origin.
Nevertheless, it is easy to misuse unsupervised clustering methods to detect disease
subtypes. These methods are designed to group data points in multidimensional space
and usually will make grouping calls even when there is no clear discontinuity in the
distribution. The robustness of the subgroups should be evaluated with varying feature-
selection methods and clustering algorithms and with cross-validation when possible.
Visual inspection of the clustering result by heatmaps is important. Testing for
associations of the clustering results with other known clinical or molecular features will
also help to evaluate the validity of subtypes.
In addition, sources of systematic noise such as tumor purity and batch effects
may drive the clustering if the biological heterogeneity (the ’signal’) is low. Such noise
!
162
could confound some of the associations found in the post hoc examinations mentioned
above. It is possible that tumor purity reflects different biology of the tumor, but it is more
likely to be associated randomness in the sampling process. Even if the tumor purity
indeed reflect different biology, it is usually caused by differences in a limited set of loci,
but in turn affects the measurement at almost all loci with tumor-specific changes.
Therefore, tumor purity must be taken into account of such studies. One approach would
be to estimate the purity, and restore the measurement in silico via deconvolution
methods.
Driver versus Passenger
The epigenome of a cancer cell is determined in part by the cell of origin for that
cancer and includes passenger hypermethylation events at genes not required in that
particular lineage. Many of the hypermethylation events do not seem to cause
transcriptional change, and reflect upstream molecular disturbance that leads to
changes at multiple loci. For example, CIMP-associated DNA methylation events are
highly correlated, with a large number of recurrent alterations that appear to be
passenger events without functional contribution to the cancer process. This high degree
of correlation precludes the straightforward use of recurrence frequency among different
tumors as a main filter criterion in the identification of functionally relevant epigenetic
driver events. Therefore, the identification of epigenetic drivers must rely more on 1) the
analysis of transcriptional consequences, 2) mutual exclusivity with other events in the
same pathway within a tumor, 3) complementary mechanisms of inactivation of the
same gene in other tumors, and most importantly, 4) functional experimental validation
of an impact of the epigenetic gene inactivation on cellular proliferation, immortality,
!
163
angiogenesis, cell death, invasion or metastasis. We have identified genes
epigenetically silenced several cancer types as part of this dissertation based on
Principle 1, and evaluated those based on Principles 2 and 3, and those genes awaits to
be tested in functional studies as described in 4.
Genetic-Epigenetic Interplay
It is clear that the cancer genome and epigenome influence each other in a
multitude of ways. They offer complementary mechanisms to achieve similar results,
such as the inactivation of tumor-suppressor genes such as BRCA1, CDKN2A, VHL,
and RB1 by either deletion or epigenetic silencing. Our finding of mutual exclusivity of
epigenetic silencing of BRCA1 and VHL with mutations in the same genes attests to
that. They can also work cooperatively, as in the case of CIMP and BRAF mutation in
colorectal cancer, where CIMP appears to create a permissive context for BRAF
mutation as early as in the precursor lesion (Hinoue et al., 2009; Yamamoto et al.,
2012).
! The genome affects the epigenome via both in cis – the DNA sequence itself that
the epigenetic modifications occur upon and in trans – the epigenetic regulators that the
genome encodes. The explosion in the number of epigenetic regulator mutations
identified in human cancer has underscored the importance of epigenetic control in
tumor suppression, although the phenotypic consequences of these mutations remain
largely uncharacterized. Our finding of a distinct DNA methylation profile associated
with SETD2 mutation contributes to the understanding of mutations in epigenetic
regulators in cancer.
!
164
Apart from the somatic changes, previous studies have shown that germline
variants of MLH1 and MSH2 can predispose to extensive somatic epigenetic silencing of
these genes, and thereby increase cancer risk (Hitchins et al., 2011; Ligtenberg et al.,
2009).! Our HNF1B study provided a low-to-medium penetrance example of genetic
variants dispose to cancer via influencing the epigenome.
Cancer epigenetics and genetics may inform each other. Genetics can shed light
on the identity of epigenetic drivers by revealing mutual exclusivity with genetic
aberrations in the same gene or pathway. Epigenetics may also provide insight into
genetic drivers in a similar fashion. In addition, the understanding of epigenetic networks
provides a framework to interpret the functional significance of lower-frequency drivers in
the same pathway. The high frequency of epigenetic regulator mutations seen in various
cancers, the hotspot nature of some mutations found, mutual exclusivity between
different mechanisms affecting the same genes/pathways, clonal analysis highlighting
convergent evolution, and validations in experimental systems all attest to the
importance of mutations in epigenetic regulators in cancer, and strengthen the concept
that disruption of epigenetic control is a common enabling characteristic of cancer cells.
165
Bibliography
Abdel-Wahab, O., Adli, M., LaFave, L.M., Gao, J., Hricik, T., Shih, A.H., Pandey, S.,
Patel, J.P., Chung, Y.R., Koche, R., et al. (2012). ASXL1 mutations promote myeloid
transformation through loss of PRC2-mediated gene repression. Cancer Cell 22, 180-
193.
Abdel-Wahab, O., Mullally, A., Hedvat, C., Garcia-Manero, G., Patel, J., Wadleigh, M.,
Malinge, S., Yao, J., Kilpivaara, O., Bhat, R., et al. (2009). Genetic characterization of
TET1, TET2, and TET3 alterations in myeloid malignancies. Blood 114, 144-147.
Aboyoun, P., Pages, H., and Lawrence, M. (R package). GenomicRanges:
Representation and manipulation of genomic intervals, pp. Version 1.8.3.
Adams, D., Altucci, L., Antonarakis, S.E., Ballesteros, J., Beck, S., Bird, A., Bock, C.,
Boehm, B., Campo, E., Caricasole, A., et al. (2012). BLUEPRINT to decode the
epigenetic signature written in blood. Nat Biotechnol 30, 224-226.
American Cancer Society (2013). Cancer Facts & Figures 2013. Atlanta: American
Cancer Society;.
Antoniou, A.C., and Easton, D.F. (2006). Risk prediction models for familial breast
cancer. Future Oncol 2, 257-274.
Auranen, A., Pukkala, E., Makinen, J., Sankila, R., Grenman, S., and Salmi, T. (1996).
Cancer incidence in the first-degree relatives of ovarian cancer patients. Br J Cancer 74,
280-284.
Baer, C., Claus, R., and Plass, C. (2013). Genome-Wide Epigenetic Regulation of
miRNAs in Cancer. Cancer Res 73, 473-477.
Baldwin, R.L., Nemeth, E., Tran, H., Shvartsman, H., Cass, I., Narod, S., and Karlan,
B.Y. (2000). BRCA1 promoter region hypermethylation in ovarian carcinoma: a
population-based study. Cancer Res 60, 5329-5333.
Baylin, S.B., Hoppener, J.W., de Bustros, A., Steenbergh, P.H., Lips, C.J., and Nelkin,
B.D. (1986). DNA methylation patterns of the calcitonin gene in human lung cancers and
lymphomas. Cancer Res 46, 2917-2922.
166
Baylin, S.B., and Jones, P.A. (2011). A decade of exploring the cancer epigenome -
biological and translational implications. Nat Rev Cancer 11, 726-734.
Bell, J.T., Pai, A.A., Pickrell, J.K., Gaffney, D.J., Pique-Regi, R., Degner, J.F., Gilad, Y.,
and Pritchard, J.K. (2011). DNA methylation patterns associate with genetic and gene
expression variation in HapMap cell lines. Genome Biol 12, R10.
Bendik, I., Schraml, P., and Ludwig, C.U. (1998). Characterization of MAST9/Hevin, a
SPARC-like protein, that is down-regulated in non-small cell lung cancer. Cancer Res
58, 626-629.
Berger, M.F., Lawrence, M.S., Demichelis, F., Drier, Y., Cibulskis, K., Sivachenko, A.Y.,
Sboner, A., Esgueva, R., Pflueger, D., Sougnez, C., et al. (2011). The genomic
complexity of primary human prostate cancer. Nature 470, 214-220.
Berman, B.P., Weisenberger, D.J., Aman, J.F., Hinoue, T., Ramjan, Z., Liu, Y.,
Noushmehr, H., Lange, C.P., van Dijk, C.M., Tollenaar, R.A., et al. (2012). Regions of
focal DNA hypermethylation and long-range hypomethylation in colorectal cancer
coincide with nuclear lamina-associated domains. Nature Genetics 44, 40-46.
Berndt, S.I., Sampson, J., Yeager, M., Jacobs, K.B., Wang, Z., Hutchinson, A., Chung,
C., Orr, N., Wacholder, S., Chatterjee, N., et al. (2011). Large-scale fine mapping of the
HNF1B locus and prostate cancer risk. Hum Mol Genet 20, 3322-3329.
Bernstein, B.E., Mikkelsen, T.S., Xie, X., Kamal, M., Huebert, D.J., Cuff, J., Fry, B.,
Meissner, A., Wernig, M., Plath, K., et al. (2006). A bivalent chromatin structure marks
key developmental genes in embryonic stem cells. Cell 125, 315-326.
Bibikova, M., Barnes, B., Tsan, C., Ho, V., Klotzle, B., Le, J.M., Delano, D., Zhang, L.,
Schroth, G.P., Gunderson, K.L., et al. (2011). High density DNA methylation array with
single CpG site resolution. Genomics 98, 288-295.
Bird, A., Taggart, M., Frommer, M., Miller, O.J., and Macleod, D. (1985). A fraction of the
mouse genome that is derived from islands of nonmethylated, CpG-rich DNA. Cell 40,
91-99.
Bolton, K.L., Ganda, C., Berchuck, A., Pharaoh, P.D., and Gayther, S.A. (2012). Role of
common genetic variants in ovarian cancer susceptibility and outcome: progress to date
from the Ovarian Cancer Association Consortium (OCAC). Journal of internal medicine
271, 366-378.
167
Bolton, K.L., Tyrer, J., Song, H., Ramus, S.J., Notaridou, M., Jones, C., Sher, T., Gentry-
Maharaj, A., Wozniak, E., Tsai, Y.Y., et al. (2010). Common variants at 19p13 are
associated with susceptibility to ovarian cancer. Nat Genet 42, 880-884.
Boyarchuk, E., Montes de Oca, R., and Almouzni, G. (2011). Cell cycle dynamics of
histone variants at the centromere, a model for chromosomal landmarks. Curr Opin Cell
Biol 23, 266-276.
Bracken, A.P., Pasini, D., Capra, M., Prosperini, E., Colli, E., and Helin, K. (2003). EZH2
is downstream of the pRB-E2F pathway, essential for proliferation and amplified in
cancer. EMBO J 22, 5323-5335.
Brauch, H., Weirich, G., Brieger, J., Glavac, D., Rodl, H., Eichinger, M., Feurer, M.,
Weidt, E., Puranakanitstha, C., Neuhaus, C., et al. (2000). VHL alterations in human
clear cell renal cell carcinoma: association with advanced tumor stage and a novel hot
spot mutation. Cancer Res 60, 1942-1948.
Campan, M., Weisenberger, D.J., Trinh, B., and Laird, P.W. (2009). MethyLight.
Methods in molecular biology 507, 325-337.
Cedar, H., and Bergman, Y. (2012). Programming of DNA methylation patterns. Annu
Rev Biochem 81, 97-117.
Challen, G.A., Sun, D., Jeong, M., Luo, M., Jelinek, J., Berg, J.S., Bock, C.,
Vasanthakumar, A., Gu, H., Xi, Y., et al. (2012). Dnmt3a is essential for hematopoietic
stem cell differentiation. Nat Genet 44, 23-31.
Cheng, K.W., Lahad, J.P., Kuo, W.L., Lapuk, A., Yamada, K., Auersperg, N., Liu, J.,
Smith-McCune, K., Lu, K.H., Fishman, D., et al. (2004). The RAB25 small GTPase
determines aggressiveness of ovarian and breast cancers. Nat Med 10, 1251-1256.
Chi, P., Allis, C.D., and Wang, G.G. (2010). Covalent histone modifications--miswritten,
misinterpreted and mis- erased in human cancers. Nat Rev Cancer 10, 457-469.
Claeskens, A., Ongenae, N., Neefs, J.M., Cheyns, P., Kaijen, P., Cools, M., and Kutoh,
E. (2000). Hevin is down-regulated in many cancers and is a negative regulator of cell
growth and proliferation. Br J Cancer 82, 1123-1130.
168
Clark, S.J. (2007). Action at a distance: epigenetic silencing of large chromosomal
regions in carcinogenesis. Hum Mol Genet 16, R88-R95.
Cloos, P.A., Christensen, J., Agger, K., Maiolica, A., Rappsilber, J., Antal, T., Hansen,
K.H., and Helin, K. (2006). The putative oncogene GASC1 demethylates tri- and
dimethylated lysine 9 on histone H3. Nature 442, 307-311.
Cohen, J. (1992). A power primer. Psychological bulletin 112, 155-159.
Cohen, N.M., Kenigsberg, E., and Tanay, A. (2011). Primate CpG islands are
maintained by heterogeneous evolutionary regimes involving minimal selection. Cell
145, 773-786.
Coolen, M.W., Stirzaker, C., Song, J.Z., Statham, A.L., Kassir, Z., Moreno, C.S., Young,
A.N., Varma, V., Speed, T.P., Cowley, M., et al. (2010). Consolidation of the cancer
genome into domains of repressive chromatin by long-range epigenetic silencing (LRES)
reduces transcriptional plasticity. Nat Cell Biol 12, 235-246.
Couronne, L., Bastard, C., and Bernard, O.A. (2012). TET2 and DNMT3A mutations in
human T-cell lymphoma. N Engl J Med 366, 95-96.
Craene, B.D., and Berx, G. (2012). Regulatory networks defining EMT during cancer
initiation and progression. Nat Rev Cancer 13, 97-110.
Crum, C.P., Drapkin, R., Miron, A., Ince, T.A., Muto, M., Kindelberger, D.W., and Lee, Y.
(2007). The distal fallopian tube: a new model for pelvic serous carcinogenesis. Curr
Opin Obstet Gynecol 19, 3-9.
Cyster, J.G. (1999). Chemokines and the homing of dendritic cells to the T cell areas of
lymphoid organs. J Exp Med 189, 447-450.
Dalgliesh, G.L., Furge, K., Greenman, C., Chen, L., Bignell, G., Butler, A., Davies, H.,
Edkins, S., Hardy, C., Latimer, C., et al. (2010). Systematic sequencing of renal
carcinoma reveals inactivation of histone modifying genes. Nature 463, 360-363.
Dallosso, A.R., Hancock, A.L., Szemes, M., Moorwood, K., Chilukamarri, L., Tsai, H.H.,
Sarkar, A., Barasch, J., Vuononvirta, R., Jones, C., et al. (2009). Frequent long-range
epigenetic silencing of protocadherin gene clusters on chromosome 5q31 in Wilms'
tumor. PLoS Genet 5, e1000745.
169
Dawson, M.A., and Kouzarides, T. (2012). Cancer epigenetics: from mechanism to
therapy. Cell 150, 12-27.
Daxinger, L., and Whitelaw, E. (2012). Understanding transgenerational epigenetic
inheritance via the gametes in mammals. Nat Rev Genet 13, 153-162.
De Carvalho, D.D., Sharma, S., You, J.S., Su, S.F., Taberlay, P.C., Kelly, T.K., Yang, X.,
Liang, G., and Jones, P.A. (2012). DNA methylation screening identifies driver
epigenetic events of cancer cell survival. Cancer Cell 21, 655-667.
De, S., and Michor, F. (2011). DNA secondary structures and epigenetic determinants of
cancer genome evolution. Nat Struct Mol Biol 18, 950-955.
Dey, A., Seshasayee, D., Noubade, R., French, D.M., Liu, J., Chaurushiya, M.S.,
Kirkpatrick, D.S., Pham, V.C., Lill, J.R., Bakalarski, C.E., et al. (2012). Loss of the tumor
suppressor BAP1 causes myeloid transformation. Science 337, 1541-1546.
Dhayalan, A., Rajavelu, A., Rathert, P., Tamas, R., Jurkowska, R.Z., Ragozin, S., and
Jeltsch, A. (2010). The Dnmt3a PWWP domain reads histone 3 lysine 36 trimethylation
and guides DNA methylation. J Biol Chem 285, 26114-26120.
Diala, E.S., and Hoffman, R.M. (1982). Hypomethylation of HeLa cell DNA and the
absence of 5-methylcytosine in SV40 and adenovirus (type 2) DNA: analysis by HPLC.
Biochem Biophys Res Commun 107, 19-26.
Dion, V., Lin, Y., Hubert, L., Jr., Waterland, R.A., and Wilson, J.H. (2008). Dnmt1
Deficiency Promotes CAG Repeat Expansion in the Mouse Germline. Hum Mol Genet 5,
5.
Dolnik, A., Engelmann, J.C., Scharfenberger-Schmeer, M., Mauch, J., Kelkenberg-
Schade, S., Haldemann, B., Fries, T., Kronke, J., Kuhn, M.W., Paschka, P., et al. (2012).
Commonly altered genomic regions in acute myeloid leukemia are enriched for somatic
mutations involved in chromatin remodeling and splicing. Blood 120, e83-92.
Dreijerink, K., Braga, E., Kuzmin, I., Geil, L., Duh, F.M., Angeloni, D., Zbar, B., Lerman,
M.I., Stanbridge, E.J., Minna, J.D., et al. (2001). The candidate tumor suppressor gene,
RASSF1A, from human chromosome 3p21.3 is involved in kidney tumorigenesis. Proc
Natl Acad Sci U S A 98, 7504-7509.
170
Eads, C.A., Danenberg, K.D., Kawakami, K., Saltz, L.B., Blake, C., Shibata, D.,
Danenberg, P.V., and Laird, P.W. (2000). MethyLight: a high-throughput assay to
measure DNA methylation. Nucleic acids research 28, E32.
Eads, C.A., Danenberg, K.D., Kawakami, K., Saltz, L.B., Danenberg, P.V., and Laird,
P.W. (1999). CpG island hypermethylation in human colorectal tumors is not associated
with DNA methyltransferase overexpression. Cancer Research 59, 2302-2306.
Easwaran, H., Johnstone, S.E., Van Neste, L., Ohm, J., Mosbruger, T., Wang, Q.,
Aryee, M.J., Joyce, P., Ahuja, N., Weisenberger, D., et al. (2012). A DNA
hypermethylation module for the stem/progenitor cell signature of cancer. Genome Res
22, 837-849.
Eden, A., Gaudet, F., Waghmare, A., and Jaenisch, R. (2003). Chromosomal instability
and tumors promoted by DNA hypomethylation. Science 300, 455.
Edlund, C.K., Lee, W.H., Li, D., Van Den Berg, D.J., and Conti, D.V. (2008). Snagger: a
user-friendly program for incorporating additional information for tagSNP selection. BMC
Bioinformatics 9, 174.
Eeles, R.A., Kote-Jarai, Z., Al Olama, A.A., Giles, G.G., Guy, M., Severi, G., Muir, K.,
Hopper, J.L., Henderson, B.E., Haiman, C.A., et al. (2009). Identification of seven new
prostate cancer susceptibility loci through a genome-wide association study. Nat Genet
41, 1116-1121.
Ehrlich, M., Gama-Sosa, M.A., Huang, L.H., Midgett, R.M., Kuo, K.C., McCune, R.A.,
and Gehrke, C. (1982). Amount and distribution of 5-methylcytosine in human DNA from
different types of tissues of cells. Nucleic Acids Res 10, 2709-2721.
Ehrlich, M., and Wang, R.Y. (1981). 5-Methylcytosine in eukaryotic DNA. Science 212,
1350-1357.
Encode-Project-Consortium, Dunham, I., Kundaje, A., Aldred, S.F., Collins, P.J., Davis,
C.A., Doyle, F., Epstein, C.B., Frietze, S., Harrow, J., et al. (2012). An integrated
encyclopedia of DNA elements in the human genome. Nature 489, 57-74.
Ernst, J., Kheradpour, P., Mikkelsen, T.S., Shoresh, N., Ward, L.D., Epstein, C.B.,
Zhang, X., Wang, L., Issner, R., Coyne, M., et al. (2011). Mapping and analysis of
chromatin state dynamics in nine human cell types. Nature 473, 43-49.
171
Ernst, T., Chase, A.J., Score, J., Hidalgo-Curtis, C.E., Bryant, C., Jones, A.V., Waghorn,
K., Zoi, K., Ross, F.M., Reiter, A., et al. (2010). Inactivating mutations of the histone
methyltransferase gene EZH2 in myeloid disorders. Nat Genet 42, 722-726.
Estecio, M.R., Gallegos, J., Vallot, C., Castoro, R.J., Chung, W., Maegawa, S., Oki, Y.,
Kondo, Y., Jelinek, J., Shen, L., et al. (2010). Genome architecture marked by
retrotransposons modulates predisposition to DNA methylation in cancer. Genome Res
20, 1369-1382.
Esteller, M., Garcia-Foncillas, J., Andion, E., Goodman, S.N., Hidalgo, O.F., Vanaclocha,
V., Baylin, S.B., and Herman, J.G. (2000a). Inactivation of the DNA-repair gene MGMT
and the clinical response of gliomas to alkylating agents. N Engl J Med 343, 1350-1354.
Esteller, M., Risques, R.A., Toyota, M., Capella, G., Moreno, V., Peinado, M.A., Baylin,
S.B., and Herman, J.G. (2001). Promoter hypermethylation of the DNA repair gene O(6)-
methylguanine-DNA methyltransferase is associated with the presence of G:C to A:T
transition mutations in p53 in human colorectal tumorigenesis. Cancer Res 61, 4689-
4692.
Esteller, M., Silva, J.M., Dominguez, G., Bonilla, F., Matias-Guiu, X., Lerma, E.,
Bussaglia, E., Prat, J., Harkes, I.C., Repasky, E.A., et al. (2000b). Promoter
hypermethylation and BRCA1 inactivation in sporadic breast and ovarian tumors. J Natl
Cancer Inst 92, 564-569.
Esteller, M., Silva, J.M., Dominguez, G., Bonilla, F., Matias-Guiu, X., Lerma, E.,
Bussaglia, E., Prat, J., Harkes, I.C., Repasky, E.A., et al. (2000c). Promoter
hypermethylation and BRCA1 inactivation in sporadic breast and ovarian tumors. J Natl
Cancer Inst 92, 564-569.
Esteller, M., Toyota, M., Sanchez-Cespedes, M., Capella, G., Peinado, M.A., Watkins,
D.N., Issa, J.P., Sidransky, D., Baylin, S.B., and Herman, J.G. (2000d). Inactivation of
the DNA repair gene O6-methylguanine-DNA methyltransferase by promoter
hypermethylation is associated with G to A mutations in K-ras in colorectal
tumorigenesis. Cancer Res 60, 2368-2371.
Fabbri, M., and Calin, G.A. (2010). Epigenetics and miRNAs in human cancer. Adv
Genet 70, 87-99.
Fan, Y., Xin, X.Y., Chen, B.L., and Ma, X. (2006). Knockdown of RAB25 expression by
RNAi inhibits growth of human epithelial ovarian cancer cells in vitro and in vivo.
Pathology 38, 561-567.
172
Feinberg, A.P., and Vogelstein, B. (1983). Hypomethylation distinguishes genes of some
human cancers from their normal counterparts. Nature 301, 89-92.
Fenouil, R., Cauchy, P., Koch, F., Descostes, N., Cabeza, J.Z., Innocenti, C., Ferrier, P.,
Spicuglia, S., Gut, M., Gut, I., et al. (2012). CpG islands and GC content dictate
nucleosome depletion in a transcription-independent manner at mammalian promoters.
Genome Res 22, 2399-2408.
Fiegl, H., Gattringer, C., Widschwendter, A., Schneitter, A., Ramoni, A., Sarlay, D.,
Gaugg, I., Goebel, G., Muller, H.M., Mueller-Holzner, E., et al. (2004). Methylated DNA
collected by tampons--a new tool to detect endometrial cancer. Cancer Epidemiol
Biomarkers Prev 13, 882-888.
Figueroa, M.E., Abdel-Wahab, O., Lu, C., Ward, P.S., Patel, J., Shih, A., Li, Y., Bhagwat,
N., Vasanthakumar, A., Fernandez, H.F., et al. (2010). Leukemic IDH1 and IDH2
mutations result in a hypermethylation phenotype, disrupt TET2 function, and impair
hematopoietic differentiation. Cancer Cell 18, 553-567.
Filippakopoulos, P., Qi, J., Picaud, S., Shen, Y., Smith, W.B., Fedorov, O., Morse, E.M.,
Keates, T., Hickman, T.T., Felletar, I., et al. (2010). Selective inhibition of BET
bromodomains. Nature 468, 1067-1073.
Filippova, G.N., Lindblom, A., Meincke, L.J., Klenova, E.M., Neiman, P.E., Collins, S.J.,
Doggett, N.A., and Lobanenkov, V.V. (1998). A widely expressed transcription factor
with multiple DNA sequence specificity, CTCF, is localized at chromosome segment
16q22.1 within one of the smallest regions of overlap for common deletions in breast and
prostate cancers. Genes Chromosomes Cancer 22, 26-36.
Filippova, G.N., Qi, C.F., Ulmer, J.E., Moore, J.M., Ward, M.D., Hu, Y.J., Loukinov, D.I.,
Pugacheva, E.M., Klenova, E.M., Grundy, P.E., et al. (2002). Tumor-associated zinc
finger mutations in the CTCF transcription factor selectively alter tts DNA-binding
specificity. Cancer Res 62, 48-52.
Fraga, M.F., Ballestar, E., Villar-Garea, A., Boix-Chornet, M., Espada, J., Schotta, G.,
Bonaldi, T., Haydon, C., Ropero, S., Petrie, K., et al. (2005). Loss of acetylation at Lys16
and trimethylation at Lys20 of histone H4 is a common hallmark of human cancer. Nat
Genet 37(4), 391-400.
French, C.A., Ramirez, C.L., Kolmakova, J., Hickman, T.T., Cameron, M.J., Thyne, M.E.,
Kutok, J.L., Toretsky, J.A., Tadavarthy, A.K., Kees, U.R., et al. (2008). BRD-NUT
173
oncoproteins: a family of closely related nuclear proteins that block epithelial
differentiation and maintain the growth of carcinoma cells. Oncogene 27, 2237-2242.
Fujimoto, A., Totoki, Y., Abe, T., Boroevich, K.A., Hosoda, F., Nguyen, H.H., Aoki, M.,
Hosono, N., Kubo, M., Miya, F., et al. (2012). Whole-genome sequencing of liver
cancers identifies etiological influences on mutation patterns and recurrent mutations in
chromatin regulators. Nat Genet 44, 760-764.
Gaffney, D.J., McVicker, G., Pai, A.A., Fondufe-Mittendorf, Y.N., Lewellen, N., Michelini,
K., Widom, J., Gilad, Y., and Pritchard, J.K. (2012). Controls of nucleosome positioning
in the human genome. PLoS Genet 8, e1003036.
Gal-Yam, E.N., Egger, G., Iniguez, L., Holster, H., Einarsson, S., Zhang, X., Lin, J.C.,
Liang, G., Jones, P.A., and Tanay, A. (2008). Frequent switching of Polycomb
repressive marks and DNA hypermethylation in the PC3 prostate cancer cell line. Proc
Natl Acad Sci U S A 105, 12979-12984.
Gayther, S.A., Batley, S.J., Linger, L., Bannister, A., Thorpe, K., Chin, S.F., Daigo, Y.,
Russell, P., Wilson, A., Sowter, H.M., et al. (2000). Mutations truncating the EP300
acetylase in human cancers. Nat Genet 24, 300-303.
Gerlinger, M., Rowan, A.J., Horswell, S., Larkin, J., Endesfelder, D., Gronroos, E.,
Martinez, P., Matthews, N., Stewart, A., Tarpey, P., et al. (2012). Intratumor
heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J
Med 366, 883-892.
Gertz, J., Varley, K.E., Reddy, T.E., Bowling, K.M., Pauli, F., Parker, S.L., Kucera, K.S.,
Willard, H.F., and Myers, R.M. (2011). Analysis of DNA methylation in a three-generation
family reveals widespread genetic influence on epigenetic regulation. PLoS genetics 7,
e1002228.
Gessi, M., Gielen, G.H., Hammes, J., Dorner, E., Muhlen, A.Z., Waha, A., and Pietsch,
T. (2013). H3.3 G34R mutations in pediatric primitive neuroectodermal tumors of central
nervous system (CNS-PNET) and pediatric glioblastomas: possible diagnostic and
therapeutic implications? J Neurooncol.
Giannoulatou, E., Yau, C., Colella, S., Ragoussis, J., and Holmes, C.C. (2008).
GenoSNP: a variational Bayes within-sample SNP genotyping algorithm that does not
require a reference population. Bioinformatics 24, 2209-2214.
174
Gilks, C.B. (2010). Molecular abnormalities in ovarian cancer subtypes other than high-
grade serous carcinoma. J Oncol 2010, 740968.
Ginno, P.A., Lott, P.L., Christensen, H.C., Korf, I., and Chedin, F. (2012). R-loop
formation is a distinctive characteristic of unmethylated human CpG island promoters.
Mol Cell 45, 814-825.
Gonzalo, S., Jaco, I., Fraga, M.F., Chen, T., Li, E., Esteller, M., and Blasco, M.A. (2006).
DNA methyltransferases control telomere length and telomere recombination in
mammalian cells. Nat Cell Biol 8, 416-424.
Goode, E.L., Chenevix-Trench, G., Song, H., Ramus, S.J., Notaridou, M., Lawrenson,
K., Widschwendter, M., Vierkant, R.A., Larson, M.C., Kjaer, S.K., et al. (2010). A
genome-wide association study identifies susceptibility loci for ovarian cancer at 2q31
and 8q24. Nat Genet 42, 874-879.
Gounaris, I., Charnock-Jones, D.S., and Brenton, J.D. (2011). Ovarian clear cell
carcinoma--bad endometriosis or bad endometrium? J Pathol 225, 157-160.
Graff, J.R., Gabrielson, E., Fujii, H., Baylin, S.B., and Herman, J.G. (2000). Methylation
patterns of the E-cadherin 5' CpG island are unstable and reflect the dynamic,
heterogeneous loss of E-cadherin expression during metastatic progression. J Biol
Chem 275, 2727-2732.
Grasso, C.S., Wu, Y.M., Robinson, D.R., Cao, X., Dhanasekaran, S.M., Khan, A.P.,
Quist, M.J., Jing, X., Lonigro, R.J., Brenner, J.C., et al. (2012). The mutational landscape
of lethal castration-resistant prostate cancer. Nature 487, 239-243.
Guan, B., Mao, T.L., Panuganti, P.K., Kuhn, E., Kurman, R.J., Maeda, D., Chen, E.,
Jeng, Y.M., Wang, T.L., and Shih Ie, M. (2011). Mutation and loss of expression of
ARID1A in uterine low-grade endometrioid carcinoma. Am J Surg Pathol 35, 625-632.
Gudmundsson, J., Sulem, P., Steinthorsdottir, V., Bergthorsson, J.T., Thorleifsson, G.,
Manolescu, A., Rafnar, T., Gudbjartsson, D., Agnarsson, B.A., Baker, A., et al. (2007).
Two variants on chromosome 17 confer prostate cancer risk, and the one in TCF2
protects against type 2 diabetes. Nat Genet 39, 977-983.
Gui, Y., Guo, G., Huang, Y., Hu, X., Tang, A., Gao, S., Wu, R., Chen, C., Li, X., Zhou, L.,
et al. (2011). Frequent mutations of chromatin remodeling genes in transitional cell
carcinoma of the bladder. Nat Genet 43, 875-878.
175
Guo, G., Wang, W., and Bradley, A. (2004). Mismatch repair genes identified using
genetic screens in Blm-deficient embryonic stem cells. Nature 429, 891-895.
Hahne, F., Durinck, S., Ivanek, R., and Mueller, A. (R package). Gviz: Plotting data and
annotation information along genomic coordinates, pp. Version 1.0.0.
Hanahan, D., and Weinberg, R.A. (2011). Hallmarks of cancer: the next generation. Cell
144, 646-674.
Hansen, K.D., Timp, W., Bravo, H.C., Sabunciyan, S., Langmead, B., McDonald, O.G.,
Wen, B., Wu, H., Liu, Y., Diep, D., et al. (2011). Increased methylation variation in
epigenetic domains across cancer types. Nature genetics 43, 768-775.
Hawkins, R.D., Hon, G.C., Lee, L.K., Ngo, Q., Lister, R., Pelizzola, M., Edsall, L.E.,
Kuan, S., Luu, Y., Klugman, S., et al. (2010). Distinct epigenomic landscapes of
pluripotent and lineage-committed human cells. Cell Stem Cell 6, 479-491.
Heaphy, C.M., de Wilde, R.F., Jiao, Y., Klein, A.P., Edil, B.H., Shi, C., Bettegowda, C.,
Rodriguez, F.J., Eberhart, C.G., Hebbar, S., et al. (2011). Altered telomeres in tumors
with ATRX and DAXX mutations. Science 333, 425.
Hegi, M.E., Diserens, A.C., Gorlia, T., Hamou, M.F., de Tribolet, N., Weller, M., Kros,
J.M., Hainfellner, J.A., Mason, W., Mariani, L., et al. (2005). MGMT gene silencing and
benefit from temozolomide in glioblastoma. N Engl J Med 352(10), 997-1003.
Herman, J.G., Latif, F., Weng, Y., Lerman, M.I., Zbar, B., Liu, S., Samid, D., Duan, D.S.,
Gnarra, J.R., Linehan, W.M., et al. (1994). Silencing of the VHL tumor-suppressor gene
by DNA methylation in renal carcinoma. Proc Natl Acad Sci U S A 91, 9700-9704.
Herman, J.G., Umar, A., Polyak, K., Graff, J.R., Ahuja, N., Issa, J.P., Markowitz, S.,
Willson, J.K., Hamilton, S.R., Kinzler, K.W., et al. (1998). Incidence and functional
consequences of hMLH1 promoter hypermethylation in colorectal carcinoma. Proc Natl
Acad Sci U S A 95, 6870-6875.
Hinoue, T., Weisenberger, D.J., Lange, C.P., Shen, H., Byun, H.M., Van Den Berg, D.,
Malik, S., Pan, F., Noushmehr, H., van Dijk, C.M., et al. (2012). Genome-scale analysis
of aberrant DNA methylation in colorectal cancer. Genome research 22, 271-282.
176
Hinoue, T., Weisenberger, D.J., Pan, F., Campan, M., Kim, M., Young, J., Whitehall,
V.L., Leggett, B.A., and Laird, P.W. (2009). Analysis of the association between CIMP
and BRAF in colorectal cancer by DNA methylation profiling. PLoS One 4, e8357.
Hitchins, M.P., Rapkins, R.W., Kwok, C.T., Srivastava, S., Wong, J.J., Khachigian, L.M.,
Polly, P., Goldblatt, J., and Ward, R.L. (2011). Dominantly inherited constitutional
epigenetic silencing of MLH1 in a cancer-affected family is linked to a single nucleotide
variant within the 5'UTR. Cancer cell 20, 200-213.
Hodgkinson, A., and Eyre-Walker, A. (2011). Variation in the mutation rate across
mammalian genomes. Nat Rev Genet 12, 756-766.
Hodis, E., Watson, I.R., Kryukov, G.V., Arold, S.T., Imielinski, M., Theurillat, J.P.,
Nickerson, E., Auclair, D., Li, L., Place, C., et al. (2012). A landscape of driver mutations
in melanoma. Cell 150, 251-263.
Holmfeldt, L., Wei, L., Diaz-Flores, E., Walsh, M., Zhang, J., Ding, L., Payne-Turner, D.,
Churchman, M., Andersson, A., Chen, S.C., et al. (2013). The genomic landscape of
hypodiploid acute lymphoblastic leukemia. Nat Genet.
Hon, G.C., Hawkins, R.D., Caballero, O.L., Lo, C., Lister, R., Pelizzola, M., Valsesia, A.,
Ye, Z., Kuan, S., Edsall, L.E., et al. (2012). Global DNA hypomethylation coupled to
repressive chromatin domain formation and gene silencing in breast cancer. Genome
Res 22, 246-258.
Horikawa, Y., Iwasaki, N., Hara, M., Furuta, H., Hinokio, Y., Cockburn, B.N., Lindner, T.,
Yamagata, K., Ogata, M., Tomonaga, O., et al. (1997). Mutation in hepatocyte nuclear
factor-1 beta gene (TCF2) associated with MODY. Nat Genet 17, 384-385.
Houseman, E.A., Christensen, B.C., Yeh, R.F., Marsit, C.J., Karagas, M.R., Wrensch,
M., Nelson, H.H., Wiemels, J., Zheng, S., Wiencke, J.K., et al. (2008). Model-based
clustering of DNA methylation array data: a recursive- partitioning algorithm for high-
dimensional data arising as a mixture of beta distributions. BMC Bioinformatics 9, 365.
Huang, K.C., Rao, P.H., Lau, C.C., Heard, E., Ng, S.K., Brown, C., Mok, S.C., Berkowitz,
R.S., and Ng, S.W. (2002). Relationship of XIST expression and responses of ovarian
cancer to chemotherapy. Mol Cancer Ther 1, 769-776.
177
Hussin, J., Sinnett, D., Casals, F., Idaghdour, Y., Bruat, V., Saillour, V., Healy, J.,
Grenier, J.C., de Malliard, T., Busche, S., et al. (2013). Rare allelic forms of PRDM9
associated with childhood leukemogenesis. Genome Res.
Hyland, E.M., Molina, H., Poorey, K., Jie, C., Xie, Z., Dai, J., Qian, J., Bekiranov, S.,
Auble, D.T., Pandey, A., et al. (2011). An evolutionarily 'young' lysine residue in histone
H3 attenuates transcriptional output in Saccharomyces cerevisiae. Genes &
development 25, 1306-1319.
Imielinski, M., Berger, A.H., Hammerman, P.S., Hernandez, B., Pugh, T.J., Hodis, E.,
Cho, J., Suh, J., Capelletti, M., Sivachenko, A., et al. (2012). Mapping the hallmarks of
lung adenocarcinoma with massively parallel sequencing. Cell 150, 1107-1120.
Inthal, A., Zeitlhofer, P., Zeginigg, M., Morak, M., Grausenburger, R., Fronkova, E.,
Fahrner, B., Mann, G., Haas, O.A., and Panzer-Grumayer, R. (2012). CREBBP HAT
domain mutations prevail in relapse cases of high hyperdiploid childhood acute
lymphoblastic leukemia. Leukemia 26, 1797-1803.
Irizarry, R.A., Ladd-Acosta, C., Wen, B., Wu, Z., Montano, C., Onyango, P., Cui, H.,
Gabo, K., Rongione, M., Webster, M., et al. (2009). The human colon cancer methylome
shows similar hypo- and hypermethylation at conserved tissue-specific CpG island
shores. Nat Genet 41, 178-186.
Ito, S., Shen, L., Dai, Q., Wu, S.C., Collins, L.B., Swenberg, J.A., He, C., and Zhang, Y.
(2011). Tet proteins can convert 5-methylcytosine to 5-formylcytosine and 5-
carboxylcytosine. Science 333, 1300-1303.
Jiao, Y., Shi, C., Edil, B.H., de Wilde, R.F., Klimstra, D.S., Maitra, A., Schulick, R.D.,
Tang, L.H., Wolfgang, C.L., Choti, M.A., et al. (2011). DAXX/ATRX, MEN1, and mTOR
pathway genes are frequently altered in pancreatic neuroendocrine tumors. Science
331, 1199-1203.
Jin, C., Zang, C., Wei, G., Cui, K., Peng, W., Zhao, K., and Felsenfeld, G. (2009).
H3.3/H2A.Z double variant-containing nucleosomes mark 'nucleosome-free regions' of
active promoters and other regulatory regions. Nat Genet 41, 941-945.
Johnson, D.G., and Dent, S.Y.R. (2013). Chromatin: Receiver and Quarterback for
Cellular Signals. Cell 152, In Press.
178
Johnson, W.E., Li, C., and Rabinovic, A. (2007). Adjusting batch effects in microarray
expression data using empirical Bayes methods. Biostatistics 8, 118-127.
Jones, D.T., Jager, N., Kool, M., Zichner, T., Hutter, B., Sultan, M., Cho, Y.J., Pugh, T.J.,
Hovestadt, V., Stutz, A.M., et al. (2012). Dissecting the genomic complexity underlying
medulloblastoma. Nature 488, 100-105.
Jones, P.A. (2012). Functions of DNA methylation: islands, start sites, gene bodies and
beyond. Nat Rev Genet 13, 484-492.
Jones, P.A., and Baylin, S.B. (2002). The fundamental role of epigenetic events in
cancer. Nat Rev Genet 3, 415-428.
Jones, P.A., and Laird, P.W. (1999). Cancer epigenetics comes of age. Nature Genetics
21, 163-167.
Jones, P.A., and Liang, G. (2009). Rethinking how DNA methylation patterns are
maintained. Nat Rev Genet 10, 805-811.
Jones, S., Wang, T.L., Shih Ie, M., Mao, T.L., Nakayama, K., Roden, R., Glas, R.,
Slamon, D., Diaz, L.A., Jr., Vogelstein, B., et al. (2010). Frequent mutations of chromatin
remodeling gene ARID1A in ovarian clear cell carcinoma. Science 330, 228-231.
Kamakaka, R.T., and Biggins, S. (2005). Histone variants: deviants? Genes Dev 19,
295-310.
Kerkel, K., Spadola, A., Yuan, E., Kosek, J., Jiang, L., Hod, E., Li, K., Murty, V.V.,
Schupf, N., Vilain, E., et al. (2008). Genomic surveys by methylation-sensitive SNP
analysis identify sequence- dependent allele-specific DNA methylation. Nat Genet 40,
904-908.
Kim, M., Trinh, B.N., Long, T.I., Oghamian, S., and Laird, P.W. (2004). Dnmt1 deficiency
leads to enhanced microsatellite instability in mouse embryonic stem cells. Nucleic acids
research 32, 5742-5749.
Kim, T.H., Abdullaev, Z.K., Smith, A.D., Ching, K.A., Loukinov, D.I., Green, R.D., Zhang,
M.Q., Lobanenkov, V.V., and Ren, B. (2007). Analysis of the vertebrate insulator protein
CTCF-binding sites in the human genome. Cell 128, 1231-1245.
179
Ko, M., Huang, Y., Jankowska, A.M., Pape, U.J., Tahiliani, M., Bandukwala, H.S., An, J.,
Lamperti, E.D., Koh, K.P., Ganetzky, R., et al. (2010). Impaired hydroxylation of 5-
methylcytosine in myeloid cancers with mutant TET2. Nature 468, 839-843.
Kobel, M., Kalloger, S.E., Boyd, N., McKinney, S., Mehl, E., Palmer, C., Leung, S.,
Bowen, N.J., Ionescu, D.N., Rajput, A., et al. (2008). Ovarian carcinoma subtypes are
different diseases: implications for biomarker studies. PLoS medicine 5, e232.
Kondo, Y., Shen, L., Cheng, A.S., Ahmed, S., Boumber, Y., Charo, C., Yamochi, T.,
Urano, T., Furukawa, K., Kwabi-Addo, B., et al. (2008). Gene silencing in cancer by
histone H3 lysine 27 trimethylation independent of promoter DNA methylation. Nat
Genet 40, 741-750.
Kote-Jarai, Z., Easton, D.F., Stanford, J.L., Ostrander, E.A., Schleutker, J., Ingles, S.A.,
Schaid, D., Thibodeau, S., Dork, T., Neal, D., et al. (2008). Multiple novel prostate
cancer predisposition loci confirmed by an international study: the PRACTICAL
Consortium. Cancer Epidemiol Biomarkers Prev 17, 2052-2061.
Kote-Jarai, Z., Olama, A.A., Giles, G.G., Severi, G., Schleutker, J., Weischer, M.,
Campa, D., Riboli, E., Key, T., Gronberg, H., et al. (2011). Seven prostate cancer
susceptibility loci identified by a multi-stage genome-wide association study. Nat Genet
43, 785-791.
Ku, M., Koche, R.P., Rheinbay, E., Mendenhall, E.M., Endoh, M., Mikkelsen, T.S.,
Presser, A., Nusbaum, C., Xie, X., Chi, A.S., et al. (2008). Genomewide analysis of
PRC1 and PRC2 occupancy identifies two classes of bivalent domains. PLoS genetics
4, e1000242.
Kurman, R.J., and Shih Ie, M. (2010a). The origin and pathogenesis of epithelial ovarian
cancer: a proposed unifying theory. Am J Surg Pathol 34, 433-443.
Kurman, R.J., and Shih Ie, M. (2010b). The origin and pathogenesis of epithelial ovarian
cancer: a proposed unifying theory. The American journal of surgical pathology 34, 433-
443.
Lai, A.Y., and Wade, P.A. (2011). Cancer biology and NuRD: a multifaceted chromatin
remodelling complex. Nature reviews Cancer 11, 588-596.
Laird, P.W. (2010a). Principles and challenges of genome-wide DNA methylation
analysis. Nat Rev Genet 11, 191-203.
180
Laird, P.W. (2010b). Principles and challenges of genomewide DNA methylation
analysis. Nature reviews Genetics 11, 191-203.
Laird, P.W., Jackson-Grusby, L., Fazeli, A., Dickinson, S.L., Jung, W.E., Li, E.,
Weinberg, R.A., and Jaenisch, R. (1995). Suppression of intestinal neoplasia by DNA
hypomethylation. Cell 81, 197-205.
Langemeijer, S.M., Kuiper, R.P., Berends, M., Knops, R., Aslanyan, M.G., Massop, M.,
Stevens-Linders, E., van Hoogen, P., van Kessel, A.G., Raymakers, R.A., et al. (2009).
Acquired mutations in TET2 are common in myelodysplastic syndromes. Nat Genet 41,
838-842.
Lawrence, M., Gentleman, R., and Carey, V. (2009). rtracklayer: an R package for
interfacing with genome browsers. Bioinformatics 25, 1841-1842.
Lax, S.F., and Kurman, R.J. (1997). A dualistic model for endometrial carcinogenesis
based on immunohistochemical and molecular genetic analyses. Verhandlungen der
Deutschen Gesellschaft fur Pathologie 81, 228-232.
Le Gallo, M., O'Hara, A.J., Rudd, M.L., Urick, M.E., Hansen, N.F., O'Neil, N.J., Price,
J.C., Zhang, S., England, B.M., Godwin, A.K., et al. (2012). Exome sequencing of serous
endometrial tumors identifies recurrent somatic mutations in chromatin-remodeling and
ubiquitin ligase complex genes. Nat Genet 44, 1310-1315.
Lee, E., Iskow, R., Yang, L., Gokcumen, O., Haseley, P., Luquette, L.J., 3rd, Lohr, J.G.,
Harris, C.C., Ding, L., Wilson, R.K., et al. (2012). Landscape of somatic
retrotransposition in human cancers. Science 337, 967-971.
Lee, J.T. (2012). Epigenetic regulation by long noncoding RNAs. Science 338, 1435-
1439.
Leek, J.T., Scharpf, R.B., Bravo, H.C., Simcha, D., Langmead, B., Johnson, W.E.,
Geman, D., Baggerly, K., and Irizarry, R.A. (2010). Tackling the widespread and critical
impact of batch effects in high-throughput data. Nat Rev Genet 11, 733-739.
Leek, J.T., and Storey, J.D. (2007). Capturing heterogeneity in gene expression studies
by surrogate variable analysis. PLoS Genet 3, 1724-1735.
181
Ley, T.J., Ding, L., Walter, M.J., McLellan, M.D., Lamprecht, T., Larson, D.E., Kandoth,
C., Payton, J.E., Baty, J., Welch, J., et al. (2010). DNMT3A mutations in acute myeloid
leukemia. The New England journal of medicine 363, 2424-2433.
Li, J., Harris, R.A., Cheung, S.W., Coarfa, C., Jeong, M., Goodell, M.A., White, L.D.,
Patel, A., Kang, S.H., Shaw, C., et al. (2012). Genomic hypomethylation in the human
germline associates with selective structural mutability in the human genome. PLoS
Genet 8, e1002692.
Li, M., Boehnke, M., and Abecasis, G.R. (2005). Joint modeling of linkage and
association: identifying SNPs responsible for a linkage signal. Am J Hum Genet 76, 934-
949.
Lichtenstein, P., Holm, N.V., Verkasalo, P.K., Iliadou, A., Kaprio, J., Koskenvuo, M.,
Pukkala, E., Skytthe, A., and Hemminki, K. (2000). Environmental and heritable factors
in the causation of cancer--analyses of cohorts of twins from Sweden, Denmark, and
Finland. N Engl J Med 343, 78-85.
Lienert, F., Wirbelauer, C., Som, I., Dean, A., Mohn, F., and Schubeler, D. (2011).
Identification of genetic elements that autonomously determine DNA methylation states.
Nature Genetics 43, 1091-1097.
Ligtenberg, M.J., Kuiper, R.P., Chan, T.L., Goossens, M., Hebeda, K.M., Voorendt, M.,
Lee, T.Y., Bodmer, D., Hoenselaar, E., Hendriks-Cornelissen, S.J., et al. (2009).
Heritable somatic methylation and inactivation of MSH2 in families with Lynch syndrome
due to deletion of the 3' exons of TACSTD1. Nat Genet 41, 112-117.
Lister, R., Pelizzola, M., Dowen, R.H., Hawkins, R.D., Hon, G., Tonti-Filippini, J., Nery,
J.R., Lee, L., Ye, Z., Ngo, Q.M., et al. (2009). Human DNA methylomes at base
resolution show widespread epigenomic differences. Nature 462, 315-322.
Liu, G., Bollig-Fischer, A., Kreike, B., van de Vijver, M.J., Abrams, J., Ethier, S.P., and
Yang, Z.Q. (2009). Genomic amplification and oncogenic properties of the GASC1
histone demethylase gene in breast cancer. Oncogene 28, 4491-4500.
Lohr, J.G., Stojanov, P., Lawrence, M.S., Auclair, D., Chapuy, B., Sougnez, C., Cruz-
Gordillo, P., Knoechel, B., Asmann, Y.W., Slager, S.L., et al. (2012). Discovery and
prioritization of somatic mutations in diffuse large B-cell lymphoma (DLBCL) by whole-
exome sequencing. Proc Natl Acad Sci U S A 109, 3879-3884.
182
Loughery, J.E., Dunne, P.D., O'Neill, K.M., Meehan, R.R., McDaid, J.R., and Walsh,
C.P. (2011). DNMT1 deficiency triggers mismatch repair defects in human cells through
depletion of repair protein levels in a process involving the DNA damage response.
Human molecular genetics 20, 3241-3255.
Love, C., Sun, Z., Jima, D., Li, G., Zhang, J., Miles, R., Richards, K.L., Dunphy, C.H.,
Choi, W.W., Srivastava, G., et al. (2012). The genetic landscape of mutations in Burkitt
lymphoma. Nat Genet 44, 1321-1325.
Lu, C., Ward, P.S., Kapoor, G.S., Rohle, D., Turcan, S., Abdel-Wahab, O., Edwards,
C.R., Khanin, R., Figueroa, M.E., Melnick, A., et al. (2012). IDH mutation impairs histone
demethylation and results in a block to cell differentiation. Nature 483, 474-478.
Meuleman, W., Peric-Hupkes, D., Kind, J., Beaudry, J.B., Pagie, L., Kellis, M., Reinders,
M., Wessels, L., and van Steensel, B. (2012). Constitutive nuclear lamina-genome
interactions are highly conserved and associated with A/T-rich sequence. Genome Res.
Miller, D.S., Blessing, J.A., Krasner, C.N., Mannel, R.S., Hanjani, P., Pearl, M.L.,
Waggoner, S.E., and Boardman, C.H. (2009). Phase II evaluation of pemetrexed in the
treatment of recurrent or persistent platinum-resistant ovarian or primary peritoneal
carcinoma: a study of the Gynecologic Oncology Group. J Clin Oncol 27, 2686-2691.
Mills, A.A. (2010). Throwing the cancer switch: reciprocal roles of polycomb and trithorax
proteins. Nat Rev Cancer 10, 669-682.
Minucci, S., and Pelicci, P.G. (2006). Histone deacetylase inhibitors and the promise of
epigenetic (and more) treatments for cancer. Nat Rev Cancer 6, 38-51.
Modena, P., Testi, M.A., Facchinetti, F., Mezzanzanica, D., Radice, M.T., Pilotti, S., and
Sozzi, G. (2003). UQCRH gene encoding mitochondrial Hinge protein is interrupted by a
translocation in a soft-tissue sarcoma and epigenetically inactivated in some cancer cell
lines. Oncogene 22, 4586-4593.
Mohn, F., Weber, M., Rebhan, M., Roloff, T.C., Richter, J., Stadler, M.B., Bibel, M., and
Schubeler, D. (2008). Lineage-specific polycomb targets and de novo DNA methylation
define restriction and potential of neuronal progenitors. Mol Cell 30, 755-766.
Mok, S.C., Chan, W.Y., Wong, K.K., Muto, M.G., and Berkowitz, R.S. (1996). SPARC,
an extracellular matrix protein with tumor-suppressing activity in human ovarian epithelial
cells. Oncogene 12, 1895-1901.
183
Monti, S., Tamayo, P., Mesirov, J., and Golub, T. (2003). Consensus Clustering: A
Resampling-Based Method for Class Discovery and Visualization of Gene Expression
Microarray Data. 52, 91-118.
Morin, R.D., Johnson, N.A., Severson, T.M., Mungall, A.J., An, J., Goya, R., Paul, J.E.,
Boyle, M., Woolcock, B.W., Kuchenbauer, F., et al. (2010). Somatic mutations altering
EZH2 (Tyr641) in follicular and diffuse large B-cell lymphomas of germinal-center origin.
Nat Genet 42, 181-185.
Morin, R.D., Mendez-Lago, M., Mungall, A.J., Goya, R., Mungall, K.L., Corbett, R.D.,
Johnson, N.A., Severson, T.M., Chiu, R., Field, M., et al. (2011). Frequent mutation of
histone-modifying genes in non-Hodgkin lymphoma. Nature 476, 298-303.
Morrissey, C., Martinez, A., Zatyka, M., Agathanggelou, A., Honorio, S., Astuti, D.,
Morgan, N.V., Moch, H., Richards, F.M., Kishida, T., et al. (2001). Epigenetic inactivation
of the RASSF1A 3p21.3 tumor suppressor gene in both clear cell and papillary renal cell
carcinoma. Cancer Res 61, 7277-7281.
Mortusewicz, O., Schermelleh, L., Walter, J., Cardoso, M.C., and Leonhardt, H. (2005).
Recruitment of DNA methyltransferase I to DNA repair sites. Proc Natl Acad Sci U S A
102, 8905-8909.
Mullighan, C.G., Zhang, J., Kasper, L.H., Lerach, S., Payne-Turner, D., Phillips, L.A.,
Heatley, S.L., Holmfeldt, L., Collins-Underwood, J.R., Ma, J., et al. (2011). CREBBP
mutations in relapsed acute lymphoblastic leukaemia. Nature 471, 235-239.
Mummert, S.K., Lobanenkov, V.A., and Feinberg, A.P. (2005). Association of
chromosome arm 16q loss with loss of imprinting of insulin-like growth factor-II in Wilms
tumor. Genes Chromosomes Cancer 43, 155-161.
Musselman, C.A., Lalonde, M.E., Cote, J., and Kutateladze, T.G. (2012). Perceiving the
epigenetic landscape through histone readers. Nat Struct Mol Biol 19, 1218-1227.
Nam, K.T., Lee, H.J., Smith, J.J., Lapierre, L.A., Kamath, V.P., Chen, X., Aronow, B.J.,
Yeatman, T.J., Bhartur, S.G., Calhoun, B.C., et al. (2010). Loss of Rab25 promotes the
development of intestinal neoplasia in mice and is associated with human colorectal
adenocarcinomas. J Clin Invest 120, 840-849.
184
Nelson, P.S., Plymate, S.R., Wang, K., True, L.D., Ware, J.L., Gan, L., Liu, A.Y., and
Hood, L. (1998). Hevin, an antiadhesive extracellular matrix protein, is down-regulated in
metastatic prostate adenocarcinoma. Cancer Res 58, 232-236.
Nikoloski, G., Langemeijer, S.M., Kuiper, R.P., Knops, R., Massop, M., Tonnissen, E.R.,
van der Heijden, A., Scheele, T.N., Vandenberghe, P., de Witte, T., et al. (2010).
Somatic mutations of the histone methyltransferase gene EZH2 in myelodysplastic
syndromes. Nat Genet 42, 665-667.
Noushmehr, H., Weisenberger, D.J., Diefes, K., Phillips, H.S., Pujara, K., Berman, B.P.,
Pan, F., Pelloski, C.E., Sulman, E.P., Bhat, K.P., et al. (2010). Identification of a CpG
island methylator phenotype that defines a distinct subgroup of glioma. Cancer Cell 17,
510-522.
Ntziachristos, P., Tsirigos, A., Van Vlierberghe, P., Nedjic, J., Trimarchi, T., Flaherty,
M.S., Ferres-Marco, D., da Ros, V., Tang, Z., Siegle, J., et al. (2012). Genetic
inactivation of the polycomb repressive complex 2 in T cell acute lymphoblastic
leukemia. Nat Med 18, 298-301.
Ohm, J.E., McGarvey, K.M., Yu, X., Cheng, L., Schuebel, K.E., Cope, L., Mohammad,
H.P., Chen, W., Daniel, V.C., Yu, W., et al. (2007). A stem cell-like chromatin pattern
may predispose tumor suppressor genes to DNA hypermethylation and heritable
silencing. Nat Genet 39, 237-242.
Okano, M., Bell, D.W., Haber, D.A., and Li, E. (1999). DNA methyltransferases Dnmt3a
and Dnmt3b are essential for de novo methylation and mammalian development. Cell
99, 247-257.
Olivier, M., Hollstein, M., and Hainaut, P. (2010). TP53 mutations in human cancers:
origins, consequences, and clinical use. Cold Spring Harb Perspect Biol 2, a001008.
Padar, A., Sathyanarayana, U.G., Suzuki, M., Maruyama, R., Hsieh, J.T., Frenkel, E.P.,
Minna, J.D., and Gazdar, A.F. (2003). Inactivation of cyclin D2 gene in prostate cancers
by aberrant promoter methylation. Clin Cancer Res 9, 4730-4734.
Papamichos-Chronakis, M., and Peterson, C.L. (2012). Chromatin and the genome
integrity network. Nat Rev Genet 14, 62-75.
185
Parsons, D.W., Li, M., Zhang, X., Jones, S., Leary, R.J., Lin, J.C., Boca, S.M., Carter,
H., Samayoa, J., Bettegowda, C., et al. (2011). The genetic landscape of the childhood
cancer medulloblastoma. Science 331, 435-439.
Pasqualucci, L., Trifonov, V., Fabbri, G., Ma, J., Rossi, D., Chiarenza, A., Wells, V.A.,
Grunn, A., Messina, M., Elliot, O., et al. (2011). Analysis of the coding genome of diffuse
large B-cell lymphoma. Nat Genet 43, 830-837.
Pearce, C.L., Templeman, C., Rossing, M.A., Lee, A., Near, A.M., Webb, P.M., Nagle,
C.M., Doherty, J.A., Cushing-Haugen, K.L., Wicklund, K.G., et al. (2012). Association
between endometriosis and risk of histological subtypes of ovarian cancer: a pooled
analysis of case-control studies. Lancet Oncol 13, 385-394.
Pena-Llopis, S., Vega-Rubin-de-Celis, S., Liao, A., Leng, N., Pavia-Jimenez, A., Wang,
S., Yamasaki, T., Zhrebker, L., Sivanand, S., Spence, P., et al. (2012). BAP1 loss
defines a new class of renal cell carcinoma. Nat Genet 44, 751-759.
Peters, I., Rehmet, K., Wilke, N., Kuczyk, M.A., Hennenlotter, J., Eilers, T., Machtens,
S., Jonas, U., and Serth, J. (2007). RASSF1A promoter methylation and expression
analysis in normal and neoplastic kidney indicates a role in early tumorigenesis. Mol
Cancer 6, 49.
Petruk, S., Sedkov, Y., Johnston, D.M., Hodgson, J.W., Black, K.L., Kovermann, S.K.,
Beck, S., Canaani, E., Brock, H.W., and Mazo, A. (2012). TrxG and PcG proteins but not
methylated histones remain associated with DNA through replication. Cell 150, 922-933.
Pharaoh, P.D.P., Tsai, Y.-Y., Ramus, S.J., Phelan, C.M., Goode, E.L., Lawrenson, K.,
Price, M., Fridley, B.L., Tyrer, J.P., Shen, H., et al. (In Press). GWAS meta-analysis and
replication identifies three novel common susceptibility loci for ovarian cancer. Nature
Genetics.
Press, J.Z., De Luca, A., Boyd, N., Young, S., Troussard, A., Ridge, Y., Kaurah, P.,
Kalloger, S.E., Blood, K.A., Smith, M., et al. (2008). Ovarian carcinomas with genetic
and epigenetic BRCA1 loss have distinct molecular abnormalities. BMC Cancer 8, 17.
Price, A.L., Patterson, N.J., Plenge, R.M., Weinblatt, M.E., Shadick, N.A., and Reich, D.
(2006). Principal components analysis corrects for stratification in genome-wide
association studies. Nat Genet 38, 904-909.
186
Prokhortchouk, A., Sansom, O., Selfridge, J., Caballero, I.M., Salozhin, S., Aithozhina,
D., Cerchietti, L., Meng, F.G., Augenlicht, L.H., Mariadason, J.M., et al. (2006). Kaiso-
deficient mice show resistance to intestinal cancer. Mol Cell Biol 26, 199-208.
Pugh, T.J., Weeraratne, S.D., Archer, T.C., Pomeranz Krummel, D.A., Auclair, D.,
Bochicchio, J., Carneiro, M.O., Carter, S.L., Cibulskis, K., Erlich, R.L., et al. (2012).
Medulloblastoma exome sequencing uncovers subtype-specific somatic mutations.
Nature 488, 106-110.
Quivoron, C., Couronne, L., Della Valle, V., Lopez, C.K., Plo, I., Wagner-Ballon, O., Do
Cruzeiro, M., Delhommeau, F., Arnulf, B., Stern, M.H., et al. (2011). TET2 inactivation
results in pleiotropic hematopoietic abnormalities in mouse and is a recurrent event
during human lymphomagenesis. Cancer Cell 20, 25-38.
Raddatz, G., Gao, Q., Bender, S., Jaenisch, R., and Lyko, F. (2012). Dnmt3a Protects
Active Chromosome Domains against Cancer-Associated Hypomethylation. PLoS Genet
8, e1003146.
Ram, O., Goren, A., Amit, I., Shoresh, N., Yosef, N., Ernst, J., Kellis, M., Gymrek, M.,
Issner, R., Coyne, M., et al. (2011). Combinatorial patterning of chromatin regulators
uncovered by genome-wide location analysis in human cells. Cell 147, 1628-1639.
Reich, M., Liefeld, T., Gould, J., Lerner, J., Tamayo, P., and Mesirov, J.P. (2006).
GenePattern 2.0. Nat Genet 38, 500-501.
Risch, H.A., Marrett, L.D., Jain, M., and Howe, G.R. (1996). Differences in risk factors for
epithelial ovarian cancer by histologic type. Results of a case-control study. Am J
Epidemiol 144, 363-372.
Robinson, G., Parker, M., Kranenburg, T.A., Lu, C., Chen, X., Ding, L., Phoenix, T.N.,
Hedlund, E., Wei, L., Zhu, X., et al. (2012). Novel mutations target distinct subgroups of
medulloblastoma. Nature 488, 43-48.
Rousseeuw, P.J. (1987). Silhouettes: A graphical aid to the interpretation and validation
of cluster analysis. Journal of Computational and Applied Mathematics 20, 53-65.
Rui, L., Emre, N.C., Kruhlak, M.J., Chung, H.J., Steidl, C., Slack, G., Wright, G.W., Lenz,
G., Ngo, V.N., Shaffer, A.L., et al. (2010). Cooperative epigenetic modulation by cancer
amplicon genes. Cancer Cell 18, 590-605.
187
Sandoval, J., and Esteller, M. (2012). Cancer epigenomics: beyond genomics. Curr Opin
Genet Dev 22, 50-55.
Sansom, O.J., Berger, J., Bishop, S.M., Hendrich, B., Bird, A., and Clarke, A.R. (2003).
Deficiency of Mbd2 suppresses intestinal tumorigenesis. Nat Genet 34, 145-147.
Schlesinger, Y., Straussman, R., Keshet, I., Farkash, S., Hecht, M., Zimmerman, J.,
Eden, E., Yakhini, Z., Ben-Shushan, E., Reubinoff, B.E., et al. (2007). Polycomb-
mediated methylation on Lys27 of histone H3 pre-marks genes for de novo methylation
in cancer. Nat Genet 39, 232-236.
Schneppenheim, R., Fruhwald, M.C., Gesk, S., Hasselblatt, M., Jeibmann, A., Kordes,
U., Kreuz, M., Leuschner, I., Martin Subero, J.I., Obser, T., et al. (2010). Germline
nonsense mutation and somatic inactivation of SMARCA4/BRG1 in a family with
rhabdoid tumor predisposition syndrome. Am J Hum Genet 86, 279-284.
Schuster-Bockler, B., and Lehner, B. (2012). Chromatin organization is a major influence
on regional mutation rates in human cancer cells. Nature 488, 504-507.
Schwartzentruber, J., Korshunov, A., Liu, X.Y., Jones, D.T., Pfaff, E., Jacob, K., Sturm,
D., Fontebasso, A.M., Quang, D.A., Tonjes, M., et al. (2012). Driver mutations in histone
H3.3 and chromatin remodelling genes in paediatric glioblastoma. Nature 482, 226-231.
Seidman, J.D., Horkayne-Szakaly, I., Haiba, M., Boice, C.R., Kurman, R.J., and Ronnett,
B.M. (2004). The histologic type and stage distribution of ovarian carcinomas of surface
epithelial origin. International journal of gynecological pathology : official journal of the
International Society of Gynecological Pathologists 23, 41-44.
Seligson, D.B., Horvath, S., Shi, T., Yu, H., Tze, S., Grunstein, M., and Kurdistani, S.K.
(2005). Global histone modification patterns predict risk of prostate cancer recurrence.
Nature 435, 1262-1266.
Sharma, S., De Carvalho, D.D., Jeong, S., Jones, P.A., and Liang, G. (2011).
Nucleosomes containing methylated DNA stabilize DNA methyltransferases 3A/3B and
ensure faithful epigenetic inheritance. PLoS genetics 7, e1001286.
Sharma, S., Yang, S.C., Hillinger, S., Zhu, L.X., Huang, M., Batra, R.K., Lin, J.F.,
Burdick, M.D., Strieter, R.M., and Dubinett, S.M. (2003). SLC/CCL21-mediated anti-
tumor responses require IFNgamma, MIG/CXCL9 and IP-10/CXCL10. Mol Cancer 2, 22.
188
Simon, C., Chagraoui, J., Krosl, J., Gendron, P., Wilhelm, B., Lemieux, S., Boucher, G.,
Chagnon, P., Drouin, S., Lambert, R., et al. (2012). A key role for EZH2 and associated
genes in mouse and human adult T-cell acute leukemia. Genes Dev 26, 651-656.
Simpkins, S.B., Bocker, T., Swisher, E.M., Mutch, D.G., Gersell, D.J., Kovatich, A.J.,
Palazzo, J.P., Fishel, R., and Goodfellow, P.J. (1999). MLH1 promoter methylation and
gene silencing is the primary cause of microsatellite instability in sporadic endometrial
cancers. Hum Mol Genet 8, 661-666.
Sirchia, S.M., Ramoscelli, L., Grati, F.R., Barbera, F., Coradini, D., Rossella, F., Porta,
G., Lesma, E., Ruggeri, A., Radice, P., et al. (2005). Loss of the inactive X chromosome
and replication of the active X in BRCA1-defective and wild-type breast cancer cells.
Cancer Res 65, 2139-2146.
Skirnisdottir, I., Seidal, T., Karlsson, M.G., and Sorbe, B. (2005). Clinical and biological
characteristics of clear cell carcinomas of the ovary in FIGO stages I-II. Int J Oncol 26,
177-183.
Smith, M.J., O'Sullivan, J., Bhaskar, S.S., Hadfield, K.D., Poke, G., Caird, J., Sharif, S.,
Eccles, D., Fitzpatrick, D., Rawluk, D., et al. (2013). Loss-of-function mutations in
SMARCE1 cause an inherited disorder of multiple spinal meningiomas. Nat Genet.
Song, H., Ramus, S.J., Tyrer, J., Bolton, K.L., Gentry-Maharaj, A., Wozniak, E., Anton-
Culver, H., Chang-Claude, J., Cramer, D.W., DiCioccio, R., et al. (2009). A genome-wide
association study identifies a new ovarian cancer susceptibility locus on 9p22.2. Nat
Genet 41, 996-1000.
Sproul, D., Kitchen, R.R., Nestor, C.E., Dixon, J.M., Sims, A.H., Harrison, D.J.,
Ramsahoye, B.H., and Meehan, R.R. (2012). Tissue of origin determines cancer-
associated CpG island promoter hypermethylation patterns. Genome Biol 13, R84.
Spurdle, A.B., Thompson, D.J., Ahmed, S., Ferguson, K., Healey, C.S., O'Mara, T.,
Walker, L.C., Montgomery, S.B., Dermitzakis, E.T., Australian National Endometrial
Cancer Study, G., et al. (2011). Genome-wide association study identifies a common
variant associated with risk of endometrial cancer. Nat Genet 43, 451-454.
Stram, D.O., Leigh Pearce, C., Bretsky, P., Freedman, M., Hirschhorn, J.N., Altshuler,
D., Kolonel, L.N., Henderson, B.E., and Thomas, D.C. (2003). Modeling and E-M
estimation of haplotype-specific relative risks from genotype data for a case-control
study of unrelated individuals. Human heredity 55, 179-190.
189
Stransky, N., Egloff, A.M., Tward, A.D., Kostic, A.D., Cibulskis, K., Sivachenko, A.,
Kryukov, G.V., Lawrence, M.S., Sougnez, C., McKenna, A., et al. (2011). The mutational
landscape of head and neck squamous cell carcinoma. Science 333, 1157-1160.
Sturm, D., Witt, H., Hovestadt, V., Khuong-Quang, D.A., Jones, D.T., Konermann, C.,
Pfaff, E., Tonjes, M., Sill, M., Bender, S., et al. (2012). Hotspot mutations in H3F3A and
IDH1 define distinct epigenetic and biological subgroups of glioblastoma. Cancer Cell
22, 425-437.
Sun, J., Zheng, S.L., Wiklund, F., Isaacs, S.D., Purcell, L.D., Gao, Z., Hsu, F.C., Kim,
S.T., Liu, W., Zhu, Y., et al. (2008). Evidence for two independent prostate cancer risk-
associated loci in the HNF1B gene at 17q12. Nat Genet 40, 1153-1155.
Tan, J., Jones, M., Koseki, H., Nakayama, M., Muntean, A.G., Maillard, I., and Hess, J.L.
(2011a). CBX8, a polycomb group protein, is essential for MLL-AF9-induced
leukemogenesis. Cancer Cell 20, 563-575.
Tan, M., Luo, H., Lee, S., Jin, F., Yang, J.S., Montellier, E., Buchou, T., Cheng, Z.,
Rousseaux, S., Rajagopal, N., et al. (2011b). Identification of 67 histone marks and
histone lysine crotonylation as a new type of histone modification. Cell 146, 1016-1028.
Tanay, A., O'Donnell, A.H., Damelin, M., and Bestor, T.H. (2007). Hyperconserved CpG
domains underlie Polycomb-binding sites. Proc Natl Acad Sci U S A 104, 5521-5526.
TCGA, C.G.A.R.N. (2008). Comprehensive genomic characterization defines human
glioblastoma genes and core pathways. Nature 455, 1061-1068.
TCGA, C.G.A.R.N. (2011). Integrated genomic analyses of ovarian carcinoma. Nature
474, 609-615.
TCGA, C.G.A.R.N. (2012a). Comprehensive genomic characterization of squamous cell
lung cancers. Nature 489, 519-525.
TCGA, C.G.A.R.N. (2012b). Comprehensive molecular characterization of human colon
and rectal cancer. Nature 487, 330-337.
TCGA, C.G.A.R.N. (2012c). Comprehensive molecular portraits of human breast
tumours. Nature 490, 61-70.
190
TCGA, C.G.A.R.N. (In Press). Integrated Genomic Analysis of Endometrioid and Serous
Uterine Cancer. Nature.
Teo, Y.Y., Inouye, M., Small, K.S., Gwilliam, R., Deloukas, P., Kwiatkowski, D.P., and
Clark, T.G. (2007). A genotype calling algorithm for the Illumina BeadArray platform.
Bioinformatics 23, 2741-2746.
Teschendorff, A.E., Menon, U., Gentry-Maharaj, A., Ramus, S.J., Weisenberger, D.J.,
Shen, H., Campan, M., Noushmehr, H., Bell, C.G., Maxwell, A.P., et al. (2010). Age-
dependent DNA methylation of genes that are suppressed in stem cells is a hallmark of
cancer. Genome research 20, 440-446.
Testa, J.R., Cheung, M., Pei, J., Below, J.E., Tan, Y., Sementino, E., Cox, N.J., Dogan,
A.U., Pass, H.I., Trusa, S., et al. (2011). Germline BAP1 mutations predispose to
malignant mesothelioma. Nat Genet 43, 1022-1025.
Thomas, G., Jacobs, K.B., Yeager, M., Kraft, P., Wacholder, S., Orr, N., Yu, K.,
Chatterjee, N., Welch, R., Hutchinson, A., et al. (2008). Multiple loci identified in a
genome-wide association study of prostate cancer. Nat Genet 40, 310-315.
Toyota, M., Ahuja, N., Ohe-Toyota, M., Herman, J.G., Baylin, S.B., and Issa, J.P. (1999).
CpG island methylator phenotype in colorectal cancer. Proc Natl Acad Sci U S A 96,
8681-8686.
Toyota, M., and Suzuki, H. (2010). Epigenetic drivers of genetic alterations. Adv Genet
70, 309-323.
Tsuchiya, A., Sakamoto, M., Yasuda, J., Chuma, M., Ohta, T., Ohki, M., Yasugi, T.,
Taketani, Y., and Hirohashi, S. (2003). Expression profiling in ovarian clear cell
carcinoma: identification of hepatocyte nuclear factor-1 beta as a molecular marker and
a possible molecular target for therapy of ovarian clear cell carcinoma. Am J Pathol 163,
2503-2512.
Turcan, S., Rohle, D., Goenka, A., Walsh, L.A., Fang, F., Yilmaz, E., Campos, C.,
Fabius, A.W., Lu, C., Ward, P.S., et al. (2012). IDH1 mutation is sufficient to establish
the glioma hypermethylator phenotype. Nature 483, 479-483.
Valouev, A., Johnson, S.M., Boyd, S.D., Smith, C.L., Fire, A.Z., and Sidow, A. (2011).
Determinants of nucleosome organization in primary human cells. Nature 474, 516-520.
191
van Haaften, G., Dalgliesh, G.L., Davies, H., Chen, L., Bignell, G., Greenman, C.,
Edkins, S., Hardy, C., O'Meara, S., Teague, J., et al. (2009). Somatic mutations of the
histone H3K27 demethylase gene UTX in human cancer. Nat Genet 41, 521-523.
Van Vlierberghe, P., Palomero, T., Khiabanian, H., Van der Meulen, J., Castillo, M., Van
Roy, N., De Moerloose, B., Philippe, J., Gonzalez-Garcia, S., Toribio, M.L., et al. (2010).
PHF6 mutations in T-cell acute lymphoblastic leukemia. Nat Genet 42, 338-342.
Van Vlierberghe, P., Patel, J., Abdel-Wahab, O., Lobry, C., Hedvat, C.V., Balbin, M.,
Nicolas, C., Payer, A.R., Fernandez, H.F., Tallman, M.S., et al. (2011). PHF6 mutations
in adult acute myeloid leukemia. Leukemia 25, 130-134.
Varambally, S., Cao, Q., Mani, R.S., Shankar, S., Wang, X., Ateeq, B., Laxman, B., Cao,
X., Jing, X., Ramnarayanan, K., et al. (2008). Genomic loss of microRNA-101 leads to
overexpression of histone methyltransferase EZH2 in cancer. Science 322, 1695-1699.
Varela, I., Tarpey, P., Raine, K., Huang, D., Ong, C.K., Stephens, P., Davies, H., Jones,
D., Lin, M.L., Teague, J., et al. (2011). Exome sequencing identifies frequent mutation of
the SWI/SNF complex gene PBRM1 in renal carcinoma. Nature 469, 539-542.
Verdugo, R.A., Deschepper, C.F., Munoz, G., Pomp, D., and Churchill, G.A. (2009).
Importance of randomization in microarray experimental designs with Illumina platforms.
Nucleic Acids Res 37, 5610-5618.
Versteege, I., Sevenet, N., Lange, J., Rousseau-Merck, M.F., Ambros, P.,
Handgretinger, R., Aurias, A., and Delattre, O. (1998). Truncating mutations of
hSNF5/INI1 in aggressive paediatric cancer. Nature 394, 203-206.
Vicari, A.P., Ait-Yahia, S., Chemin, K., Mueller, A., Zlotnik, A., and Caux, C. (2000).
Antitumor effects of the mouse chemokine 6Ckine/SLC through angiostatic and
immunological mechanisms. J Immunol 165, 1992-2000.
Wagner, E.J., and Carpenter, P.B. (2012). Understanding the language of Lys36
methylation at histone H3. Nature reviews Molecular cell biology 13, 115-126.
Wang, R.Y., Kuo, K.C., Gehrke, C.W., Huang, L.H., and Ehrlich, M. (1982). Heat- and
alkali-induced deamination of 5-methylcytosine and cytosine residues in DNA. Biochim
Biophys Acta 697, 371-377.
192
Wang, Z.Y., and Chen, Z. (2008). Acute promyelocytic leukemia: from highly fatal to
highly curable. Blood 111, 2505-2515.
Weisenberger, D.J., Campan, M., Long, T.I., Kim, M., Woods, C., Fiala, E., Ehrlich, M.,
and Laird, P.W. (2005). Analysis of repetitive element DNA methylation by MethyLight.
Nucleic acids research 33, 6823-6836.
Weisenberger, D.J., Siegmund, K.D., Campan, M., Young, J., Long, T.I., Faasse, M.A.,
Kang, G.H., Widschwendter, M., Weener, D., Buchanan, D., et al. (2006). CpG island
methylator phenotype underlies sporadic microsatellite instability and is tightly
associated with BRAF mutation in colorectal cancer. Nature Genetics 38, 787-793.
Weissmann, S., Alpermann, T., Grossmann, V., Kowarsch, A., Nadarajah, N., Eder, C.,
Dicker, F., Fasan, A., Haferlach, C., Haferlach, T., et al. (2012). Landscape of TET2
mutations in acute myeloid leukemia. Leukemia 26, 934-942.
Widschwendter, M., Fiegl, H., Egle, D., Mueller-Holzner, E., Spizzo, G., Marth, C.,
Weisenberger, D.J., Campan, M., Young, J., Jacobs, I., et al. (2007a). Epigenetic stem
cell signature in cancer. Nat Genet 39, 157-158.
Widschwendter, M., Fiegl, H., Egle, D., Mueller-Holzner, E., Spizzo, G., Marth, C.,
Weisenberger, D.J., Campan, M., Young, J., Jacobs, I., et al. (2007b). Epigenetic stem
cell signature in cancer. Nature Genetics 39, 157-158.
Wiegand, K.C., Shah, S.P., Al-Agha, O.M., Zhao, Y., Tse, K., Zeng, T., Senz, J.,
McConechy, M.K., Anglesio, M.S., Kalloger, S.E., et al. (2010). ARID1A mutations in
endometriosis-associated ovarian carcinomas. N Engl J Med 363, 1532-1543.
Wiesner, T., Obenauf, A.C., Murali, R., Fried, I., Griewank, K.G., Ulz, P., Windpassinger,
C., Wackernagel, W., Loy, S., Wolf, I., et al. (2011). Germline mutations in BAP1
predispose to melanocytic tumors. Nat Genet 43, 1018-1021.
Williams, K., Christensen, J., and Helin, K. (2012). DNA methylation: TET proteins-
guardians of CpG islands? EMBO Rep 13, 28-35.
Wilson, B.G., and Roberts, C.W. (2011). SWI/SNF nucleosome remodellers and cancer.
Nat Rev Cancer 11, 481-492.
193
Winckler, W., Weedon, M.N., Graham, R.R., McCarroll, S.A., Purcell, S., Almgren, P.,
Tuomi, T., Gaudet, D., Bostrom, K.B., Walker, M., et al. (2007). Evaluation of common
variants in the six known maturity-onset diabetes of the young (MODY) genes for
association with type 2 diabetes. Diabetes 56, 685-693.
Wu, G., Broniscer, A., McEachron, T.A., Lu, C., Paugh, B.S., Becksfort, J., Qu, C., Ding,
L., Huether, R., Parker, M., et al. (2012). Somatic histone H3 alterations in pediatric
diffuse intrinsic pontine gliomas and non-brainstem glioblastomas. Nat Genet 44, 251-
253.
Xu, K., Wu, Z.J., Groner, A.C., He, H.H., Cai, C., Lis, R.T., Wu, X., Stack, E.C., Loda,
M., Liu, T., et al. (2012). EZH2 oncogenic activity in castration-resistant prostate cancer
cells is Polycomb-independent. Science 338, 1465-1469.
Xu, W., Yang, H., Liu, Y., Yang, Y., Wang, P., Kim, S.H., Ito, S., Yang, C., Xiao, M.T.,
Liu, L.X., et al. (2011). Oncometabolite 2-hydroxyglutarate is a competitive inhibitor of
alpha-ketoglutarate-dependent dioxygenases. Cancer Cell 19, 17-30.
Yamamoto, E., Suzuki, H., Yamano, H.O., Maruyama, R., Nojima, M., Kamimae, S.,
Sawada, T., Ashida, M., Yoshikawa, K., Kimura, T., et al. (2012). Molecular dissection of
premalignant colorectal lesions reveals early onset of the CpG island methylator
phenotype. Am J Pathol 181, 1847-1861.
Yamashita, K., Dai, T., Dai, Y., Yamamoto, F., and Perucho, M. (2003). Genetics
supersedes epigenetics in colon cancer phenotype. Cancer Cell 4, 121-131.
Yamazaki, J., Estecio, M.R., Lu, Y., Long, H., Malouf, G.G., Graber, D., Huo, Y.,
Ramagli, L., Liang, S., Kornblau, S.M., et al. (2013). The epigenome of AML stem and
progenitor cells. Epigenetics 8, 92-104.
Yan, X.J., Xu, J., Gu, Z.H., Pan, C.M., Lu, G., Shen, Y., Shi, J.Y., Zhu, Y.M., Tang, L.,
Zhang, X.W., et al. (2011). Exome sequencing identifies somatic mutations of DNA
methyltransferase gene DNMT3A in acute monocytic leukemia. Nature genetics 43, 309-
315.
Yang, X.J. (2004). The diverse superfamily of lysine acetyltransferases and their roles in
leukemia and other diseases. Nucleic Acids Res 32, 959-976.
Ying, J., Li, H., Seng, T.J., Langford, C., Srivastava, G., Tsao, S.W., Putti, T., Murray, P.,
Chan, A.T., and Tao, Q. (2006). Functional epigenetics identifies a protocadherin
194
PCDH10 as a candidate tumor suppressor for nasopharyngeal, esophageal and multiple
other carcinomas with frequent methylation. Oncogene 25, 1070-1080.
Yokoyama, A., and Cleary, M.L. (2008). Menin critically links MLL proteins with LEDGF
on cancer-associated target genes. Cancer Cell 14, 36-46.
Yoon, J.H., Dammann, R., and Pfeifer, G.P. (2001). Hypermethylation of the CpG island
of the RASSF1A gene in ovarian and renal cell carcinomas. Int J Cancer 94, 212-217.
You, J.S., and Jones, P.A. (2012). Cancer genetics and epigenetics: two sides of the
same coin? Cancer Cell 22, 9-20.
Zbar, B., Brauch, H., Talmadge, C., and Linehan, M. (1987). Loss of alleles of loci on the
short arm of chromosome 3 in renal cell carcinoma. Nature 327, 721-724.
Zbar, B., Kishida, T., Chen, F., Schmidt, L., Maher, E.R., Richards, F.M., Crossey, P.A.,
Webster, A.R., Affara, N.A., Ferguson-Smith, M.A., et al. (1996). Germline mutations in
the Von Hippel-Lindau disease (VHL) gene in families from North America, Europe, and
Japan. Hum Mutat 8, 348-357.
Zeisig, B.B., Kulasekararaj, A.G., Mufti, G.J., and So, C.W. (2012). SnapShot: Acute
myeloid leukemia. Cancer Cell 22, 698-698 e691.
Zhang, J., Ding, L., Holmfeldt, L., Wu, G., Heatley, S.L., Payne-Turner, D., Easton, J.,
Chen, X., Wang, J., Rusch, M., et al. (2012). The genetic basis of early T-cell precursor
acute lymphoblastic leukaemia. Nature 481, 157-163.
Zhu, J., Adli, M., Zou, J.Y., Verstappen, G., Coyne, M., Zhang, X., Durham, T., Miri, M.,
Deshpande, V., De Jager, P.L., et al. (2013). Genome-wide Chromatin State Transitions
Associated with Developmental and Environmental Cues. Cell 152, 642-654.
Zullo, J.M., Demarco, I.A., Pique-Regi, R., Gaffney, D.J., Epstein, C.B., Spooner, C.J.,
Luperchio, T.R., Bernstein, B.E., Pritchard, J.K., Reddy, K.L., et al. (2012). DNA
sequence-dependent compartmentalization and silencing of chromatin at the nuclear
lamina. Cell 149, 1474-1487.
195
Appendix A: A list of the 168 candidate epigenetically silenced
genes in serous ovarian cancer
Official Gene
Symbol
ProbeID Sample Mean
Beta Values
of Fallopian
Tube
Beta Value
Difference
Log2
(Fold
Change)
Spearman
's Rho
RAB25 cg19580810 0.33 0.50 4.57 -0.89
LYPLAL1 cg02665570 0.28 0.46 2.77 -0.86
ZNF597 cg24333473 0.13 0.60 2.11 -0.77
VTCN1 cg22424746 0.43 0.30 4.51 -0.76
VSIG2 cg02082342 0.37 0.43 2.58 -0.75
BANK1 cg25023994 0.33 0.41 1.96 -0.74
C3 cg17612991 0.35 0.43 5.07 -0.74
DNALI1 cg21488617 0.22 0.38 3.11 -0.72
LDHD cg03991512 0.15 0.55 1.31 -0.70
UBB cg06537829 0.16 0.48 3.29 -0.69
LTC4S cg11394785 0.08 0.66 1.98 -0.68
SULT1C4 cg17966192 0.19 0.45 1.09 -0.67
NAP1L5 cg12759554 0.49 0.37 1.93 -0.67
CFTR cg25509184 0.17 0.71 2.26 -0.65
CMBL cg11882252 0.10 0.70 2.41 -0.64
TMEM173 cg16983159 0.10 0.40 2.90 -0.63
EYA4 cg21296676 0.35 0.47 2.46 -0.62
S100A16 cg23851011 0.45 0.35 1.68 -0.62
LRG1 cg24926276 0.29 0.26 3.62 -0.62
ALDOC cg06367117 0.13 0.57 1.25 -0.61
CDO1 cg07644368 0.40 0.44 2.08 -0.61
ZNF671 cg19246110 0.11 0.75 1.45 -0.59
ZNF502 cg21672276 0.11 0.21 2.18 -0.58
TSPYL5 cg15747595 0.22 0.47 3.80 -0.58
MT1E cg20083730 0.06 0.31 2.42 -0.57
TRIM4 cg01626227 0.06 0.32 1.92 -0.56
CFI cg12243271 0.28 0.23 3.16 -0.56
DNAJB13 cg19692710 0.36 0.31 1.87 -0.56
AQP9 cg11098259 0.23 0.39 2.98 -0.56
THY1 cg12508624 0.12 0.64 1.21 -0.55
CDH16 cg14221831 0.43 0.36 1.66 -0.55
KLK11 cg09702010 0.20 0.38 4.00 -0.55
CLIP3 cg06432655 0.23 0.44 1.49 -0.55
NUPR1 cg05590982 0.18 0.44 2.68 -0.54
EHF cg18414381 0.27 0.14 2.79 -0.54
UQCRH cg21576698 0.42 0.49 3.30 -0.54
196
LCN12 cg19534945 0.40 0.46 2.29 -0.54
TMEM71 cg20955688 0.16 0.53 1.97 -0.54
TMEM140 cg06456031 0.20 0.33 1.32 -0.54
SCG5 cg15787039 0.14 0.36 2.43 -0.54
KLK10 cg06130787 0.15 0.21 1.81 -0.54
WT1 cg16463460 0.18 0.25 4.79 -0.53
SCARA3 cg26847866 0.15 0.22 2.51 -0.53
APOL6 cg19853703 0.07 0.14 1.92 -0.52
SLAIN1 cg08504583 0.38 0.45 2.25 -0.52
PCDHB5 cg03349953 0.17 0.51 2.03 -0.51
HSPB2 cg13210534 0.12 0.33 1.65 -0.51
CHI3L2 cg26366091 0.22 0.39 1.09 -0.51
SLC15A2 cg10523671 0.36 0.53 2.43 -0.50
RBP1 cg13099330 0.09 0.58 3.53 -0.50
CDH6 cg10919204 0.34 0.38 1.60 -0.50
LRRC34 cg24777454 0.10 0.50 3.13 -0.50
SPDEF cg07705908 0.34 0.21 1.66 -0.50
PDLIM4 cg01305625 0.35 0.23 1.61 -0.49
RERG cg19205533 0.31 0.38 4.03 -0.49
EFS cg07197059 0.22 0.47 1.06 -0.49
HSPA1A cg05920090 0.04 0.18 3.93 -0.49
HOXB8 cg15539420 0.33 0.38 1.13 -0.49
MFAP4 cg09606564 0.19 0.45 3.35 -0.48
AMT cg25021247 0.38 0.35 1.90 -0.48
CRAT cg26805528 0.17 0.55 1.64 -0.48
PCDHB2 cg02260587 0.16 0.45 1.80 -0.48
CRISPLD1 cg01410472 0.05 0.31 2.88 -0.48
HOXB2 cg09313705 0.31 0.32 3.24 -0.48
FXYD1 cg27461196 0.36 0.31 2.99 -0.47
APOBEC3G cg26022401 0.09 0.24 3.43 -0.47
HP cg06172871 0.39 0.30 3.00 -0.46
ZNF300 cg19014419 0.12 0.24 1.97 -0.46
TRIM22 cg12461141 0.32 0.39 3.31 -0.46
RIPK3 cg20822579 0.21 0.36 1.54 -0.45
HLA-DMA cg14833385 0.12 0.11 3.47 -0.45
DDR2 cg22740835 0.15 0.69 2.42 -0.45
STAT5A cg03001305 0.22 0.33 1.24 -0.45
WDR69 cg14329157 0.26 0.45 5.30 -0.45
PTGDS cg11546621 0.41 0.37 2.62 -0.45
DDO cg20011134 0.34 0.46 1.34 -0.44
GYPC cg13901526 0.22 0.54 2.27 -0.44
PAM cg20131596 0.20 0.28 2.20 -0.44
CRYAB cg15227610 0.33 0.40 2.90 -0.44
FADS2 cg06781209 0.17 0.47 1.20 -0.44
FCGRT cg15528736 0.31 0.47 1.92 -0.43
197
ARSE cg11964613 0.35 0.48 1.32 -0.43
RNASE1 cg05958352 0.41 0.35 1.62 -0.43
AGT cg19125606 0.43 0.44 1.95 -0.43
TBX2 cg12163132 0.19 0.12 2.36 -0.43
PKIA cg04689061 0.17 0.37 1.22 -0.42
THNSL2 cg07952391 0.08 0.27 2.05 -0.41
FOXJ1 cg24164563 0.08 0.19 3.66 -0.41
CPNE8 cg23495733 0.13 0.72 1.68 -0.41
CYBRD1 cg10731149 0.12 0.14 1.64 -0.41
IL20RA cg22487322 0.43 0.35 3.08 -0.41
SPATA18 cg09022993 0.10 0.51 4.31 -0.40
PLSCR4 cg24315815 0.27 0.22 2.99 -0.40
C1S cg05538432 0.35 0.28 3.59 -0.40
VNN2 cg10044101 0.45 0.34 1.79 -0.40
TMEM101 cg12259256 0.09 0.59 2.92 -0.39
FOLR1 cg03699566 0.30 0.23 3.56 -0.39
VAMP5 cg11108890 0.30 0.54 1.36 -0.39
GSTM2 cg16670497 0.04 0.11 1.69 -0.39
PART1 cg09712066 0.37 0.12 3.50 -0.38
PNOC cg03642518 0.24 0.26 3.26 -0.38
SEMA3E cg18464137 0.14 0.41 1.10 -0.38
SERPINA3 cg06190732 0.31 0.44 4.76 -0.38
TRIM59 cg10273210 0.09 0.32 1.18 -0.38
LIMS3 cg18879041 0.20 0.13 3.81 -0.38
CYP4B1 cg23440155 0.16 0.16 4.08 -0.38
SPAG6 cg06908778 0.20 0.62 4.94 -0.38
OVGP1 cg09558502 0.13 0.10 6.66 -0.38
SERPINB1 cg06148264 0.16 0.13 3.02 -0.38
GIMAP2 cg25918245 0.24 0.24 1.96 -0.37
CLEC11A cg13152535 0.26 0.32 1.12 -0.37
IQGAP2 cg02387679 0.14 0.49 1.08 -0.37
WIT1 cg19718882 0.10 0.13 2.10 -0.37
KIAA1324 cg16797831 0.13 0.40 3.78 -0.37
ANGPTL1 cg07044282 0.39 0.20 2.64 -0.36
MCAM cg21096399 0.29 0.34 1.00 -0.36
CRIP1 cg02000005 0.18 0.37 2.65 -0.36
SLC47A2 cg24743310 0.36 0.16 3.66 -0.36
GNB4 cg17483510 0.10 0.36 1.22 -0.36
GAS2L2 cg24922045 0.30 0.17 2.26 -0.35
ZMYND12 cg06346081 0.16 0.14 2.36 -0.35
ALDH3B1 cg07730301 0.16 0.13 1.86 -0.35
SLC44A4 cg07363637 0.38 0.16 4.05 -0.35
NUAK1 cg23555120 0.48 0.36 1.63 -0.35
HOXB5 cg01405107 0.13 0.48 1.89 -0.35
LY75 cg23995753 0.23 0.32 2.68 -0.35
198
CXCR7 cg03626672 0.17 0.16 1.92 -0.35
PLAT cg12091331 0.14 0.19 3.65 -0.34
CTSO cg11754095 0.10 0.11 2.60 -0.34
ZNF655 cg13636404 0.05 0.12 1.99 -0.34
CAMK2N1 cg08398233 0.17 0.29 1.94 -0.34
BRCA1 cg04658354 0.06 0.46 1.17 -0.34
ANXA6 cg21623671 0.10 0.10 1.61 -0.34
GIPC2 cg09107315 0.32 0.44 1.38 -0.33
IL1R2 cg20340242 0.35 0.35 1.63 -0.33
IGF1 cg01305421 0.43 0.35 2.23 -0.33
KCTD14 cg17272843 0.08 0.21 2.67 -0.33
STEAP2 cg27626102 0.13 0.23 1.92 -0.33
NPDC1 cg26581729 0.38 0.42 3.19 -0.32
FBLN2 cg00201234 0.30 0.40 1.14 -0.32
H1F0 cg07141002 0.25 0.16 1.64 -0.32
GCNT3 cg06817269 0.37 0.13 2.50 -0.32
NDN cg12532169 0.48 0.30 3.90 -0.32
TRIM2 cg12793610 0.23 0.20 2.04 -0.31
CPXM2 cg09619146 0.23 0.43 2.02 -0.31
HNF1B cg12788467 0.19 0.62 2.31 -0.31
CTSS cg08578023 0.15 0.21 2.39 -0.31
NME5 cg25507001 0.11 0.11 3.48 -0.31
SLC16A5 cg09300114 0.23 0.14 1.77 -0.31
PEG3 cg18668753 0.41 0.34 2.19 -0.30
BLNK cg16779976 0.25 0.20 2.22 -0.30
RARRES2 cg17279839 0.12 0.50 3.14 -0.30
SPARCL1 cg19466563 0.16 0.62 3.47 -0.28
CBX7 cg23124451 0.33 0.50 2.17 -0.27
CCDC65 cg02620769 0.04 0.39 3.98 -0.27
APH1B cg17207590 0.26 0.38 1.98 -0.27
TSC22D3 cg00404599 0.38 0.38 1.67 -0.26
CCL21 cg27443224 0.35 0.45 3.38 -0.25
MRGPRF cg22933847 0.34 0.33 1.95 -0.25
HSPA2 cg16319578 0.16 0.47 2.77 -0.24
PENK cg24645221 0.07 0.47 1.85 -0.24
LONRF2 cg12232463 0.36 0.47 3.29 -0.23
SERPING1 cg09061733 0.18 0.32 2.19 -0.22
ALDH1A3 cg21359747 0.23 0.70 1.92 -0.22
MGP cg00431549 0.30 0.31 2.99 -0.22
CYYR1 cg10238818 0.07 0.48 2.45 -0.21
TCTEX1D1 cg24110050 0.25 0.56 3.80 -0.21
AIF1 cg21440587 0.17 0.46 1.76 -0.21
199
Appendix B: OCAC Study Sites
Supplementary Table S1. Distribution of cases and controls by study site.
Geographic Region St udy Design All Serous Mucinous Endometrioid Clear Cell Brenner Other
Australia Ovarian Cancer Study & Australia Cancer Study (Ovarian
Cancer) (AUS)
Australia Population-based/case-control 1011 949 592 40 123 64 40 90
Bavarian Ovarian Cancer Cases and Controls (BAV) Sout heast G ermany Population-based/case-control 143 93 56 8 13 6 1 9
Belgium Ovarian Cancer Study (BEL)
Belgium, University Hospital
Leuven
Hospital-based/case-control 1352 277 195 25 22 23 2 10
Diseases of the Ovary and their Evaluation (DOV)
USA: 13 counties in western
Wasthington state
Population-based/case-control 1606 990 576 27 161 75 151 0
Germany Ovarian Cancer S t udy (GER)
Germany: t wo geographical
regions in the states of
BadenWürttemberg and
Rhineland-Palatinate in
southern Germany
Population-based/case-control 413 192 96 22 21 6 1 46
Gilda Radner Familial Ovarian Cancer Regist ry (GRR)* USA Familial cancer/case only 0 115 76 5 19 11 3 1
Hawaii Ovarian Cancer Study (HAW) USA: Hawaii Population-based/case-control 601 266 130 27 60 35 6 8
Hannover-Jena Ovarian Cancer Study (HJO) Germany Hospital-based/case-control 274 273 142 9 26 4 38 54
Hannover-Minsk Ovarian Cancer Study (HMO) Belarus Case-control 140 144 50 7 12 1 0 74
Helsinki Ovarian Cancer Study (HOC) Helsinki, Finland Case-control 447 218 113 45 28 14 0 18
Hormones and Ovarian Cancer Prediction (HOP)
Western Pennsy, Northeastern
Ohio, Western New York
Population-based/case-control 1501 682 388 32 90 43 50 79
DNA-Specimen in Gynecologic Oncologic Malignancies (HSK)* Germany Case only 0 146 109 1 16 0 3 17
Hospital-based Epidemiologic Research Program at Aichi Cancer
Center (JPN)
Japan: Nagoya City Case-control 81 66 32 3 7 17 4 3
Women's Cancer Research Institute - Cedars-Sinai Medical Center
(LAX)*
USA: Southern California Case only 0 330 248 15 26 13 27 1
Danish Malignant Ovarian Tumor Study (MAL) Denmark Population-based/case-control 829 440 272 42 54 33 0 39
Malaysia Ovarian Cancer Study (MAS) Malaysia Hospital-based/case-control 106 106 44 17 25 12 1 7
Mayo Clinic Ovarian Cancer Case Control Study (MAY)
USA: North Central
(MN, SD, ND, IL, IA, WI)
Clinic-based/ case-control 753 708 515 18 97 34 0 44
Melbourne Collaborative Cohort Study (MCC) Melbourne, Australia Cohort/Nested case-control 68 64 34 7 7 6 6 4
MD Anderson Ovarian Cancer Study (MDA) USA: Texas Hospital-based/case-control 385 323 194 29 29 4 1 66
Memorial Sloan Kettering Cancer Center Gynecology Tissue Bank
(MSK)
USA: New York City Case-control 697 556 450 0 25 22 0 59
North Carolina Ovarian Cancer Study (NCO)
USA: Central and eastern
North Carolina (48 counties)
Population-based/case-control 984 850 480 43 130 85 112 0
New England-based Case-Control Study of Ovarian Cancer (NEC)
USA: New Hampshire and
Eastern Massachusetts
Population-based/case-control 1049 697 397 44 131 97 0 28
Nurses' Health Study (NHS) USA
Popluation-based/nested case-
control
429 127 68 7 14 6 13 19
New Jersey Ovarian Cancer Study (NJO) USA: New Jersey (six counties) Case-control 194 190 110 7 30 23 0 19
University of Bergen Norway Study (NOR) Norway Case-control 371 237 136 15 27 13 0 46
Nijmegen Polygene Study & Nijmegen Biomedical Study (NTH) Eastern part of the Netherlands Case-control 323 263 119 34 67 21 9 13
Oregon Ovarian Cancer Registry (ORE)* Portland, Oregon Case only 0 59 41 4 4 4 0 6
Ovarian Cancer in Alberta and British Columbia Study (OVA)
Alberta and British Columbia,
Canada
Case-control 810 688 370 29 114 73 12 90
Poland Ovarian Cancer Study (POC)
Poland: Szczecin, Poznan,
Opole, Rzeszów
Case-control 417 423 200 33 39 9 61 81
NCI Ovarian Case-Control Study in Poland (POL) Poland, Warszaw and Lodz Population-based/case-control 223 236 106 17 37 10 25 41
Pelvic Mass Study (PVD)* Denmark Population-based/case-control 0 172 130 11 14 8 6 3
Royal Marsden Hospital Case Series (RMH)* UK: London Hospital based/case only 0 151 52 16 29 17 0 37
UK Studies of Epidemiology and Risk Factors in Cancer Heredity
Ovarian Cancer Study (SEA)
UK: East Anglia and West
Midlands
Population-based/case-control 6067 1395 581 145 231 147 9 282
Sout hampt on Ovarian Cancer St udy (SOC)* United Kingdom, Wessex region
Case only/ hospital-based
0 274 105 34 64 11 7 53
Scot t ish Randomised T rial in Ovarian Cancer (SRO)*
Coordinated through clinical
trials unit, Glasgow UK from
patients recruited worldwide
Case only from clinical trial 0 159 93 3 17 9 25 12
Genet ic Epidemiology of Ovarian Cancer (S T A)
USA: Six counties in the San
Francisco Bay area
Population-based/case-control 404 282 174 19 38 22 1 28
Shanghai Women's Health Study (SWH) Shanghai, China Cohort/nested case-control 891 135 0 0 0 0 0 135
Familial Ovarian Tumor Study (TOR) Canada: Province of Ontatio Population-based 443 559 341 39 132 34 0 13
UC Irvine Ovarian Cancer Study (UCI)
USA: Southern California
(Orange and San-Diego, Imperial
Counties)
Population-based/case-control 425 331 198 24 58 29 2 20
UK Ovarian Cancer Population Study (UKO)
United Kingdom (England,
Wales and Northern Ireland)
Population-based/case-control 1123 718 357 76 116 68 55 46
UK Familial Ovarian Cancer Registry (UKR)* UK: National Case only/ Familial Register 0 48 23 3 6 2 0 14
Los Angeles County Case-Control Studies of Ovarian Cancer (USC) Los Angeles County Population-based/case-control 1370 978 614 63 124 58 26 93
Warsaw Ovarian Cancer Study (WOC)
Poland: Warsaw and central
Poland
Case-control 204 202 132 8 20 17 1 24
Total 26134 16111 9139 1053 2303 1186 698 1732
* Case only study. For our analyses, GRR was merged with HOP, HSK with GER, LAX with USC, ORE with DOV, PVD with MAL, and RMH, SOC, SRO, and UKR with UKO.
Sit e
No. of
controls
Invasive Cases
Abstract (if available)
Abstract
Cancer arises as a consequence of cumulative disruptions to cellular growth control, with Darwinian selection for those heritable changes which provide the greatest clonal advantage. These traits can be acquired and stably maintained by either genetic or epigenetic means. Alterations in the genome and epigenome could influence each other and cooperate to promote oncogenic transformation. Disruption of epigenomic control is pervasive in malignancy, and can be classified as an enabling characteristic of cancer cells, akin to genome instability and mutation. We examined epigenetic profiles of several human cancers, including ovarian, endometrial and clear cell renal cell carcinoma (ccRCC) in the context of other genomic alterations, as part of the Cancer Genome Atlas (TCGA) project. We found varying degrees of disease heterogeneity among tumors of the various cancer types studied. We found that endometrial cancer comprises several distinct molecular groups, with a serous-like subtype that is similar to serous ovarian cancer and basal-like breast cancer. We also identified important or potentially important epigenetically silenced genes and studied their clinical implications. BRCA1 is epigenetically silenced in 12% of serous ovarian cancer cases, and mutually exclusive with BRCA1/2 mutations. This epigenetic silencing is associated with worse prognosis. VHL epigenetic silencing in ccRCC is mutually exclusive with VHL mutation. Finally, we hypothesized that genetic variants in one gene that we found to be epigenetically silenced in serous ovarian cancer and differentially methylated among different ovarian cancer subtypes, HNF1B, would be associated with ovarian cancer risk in a subtype-specific way. We comprehensively mapped variation in HNF1B with respect to EOC risk. Different SNPs were associated with invasive serous (rs7405776 OR=1.13, p=3.1x10⁻¹⁰) and clear cell (rs11651755 OR=0.77, p=1.6x10⁻⁸) ovarian cancers. Risk alleles for the serous subtype were associated with higher HNF1B promoter methylation in these tumors. Unmethylated, expressed HNF1B, primarily present in clear cell tumors, coincided with a CpG Island Methylator Phenotype (CIMP) affecting numerous other promoters throughout the genome. Different variants in HNF1B are associated with risk of serous and clear cell ovarian cancers
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Integrated genomic & epigenomic analyses of glioblastoma multiforme: Methods development and application
PDF
POLR2B and its contributions to cancer cell growth
PDF
Identification and characterization of cancer-associated enhancers
PDF
Effects of chromatin regulators during carcinogenesis
PDF
Understanding and controlling mitotic errors leading to aneuploidy in early ovarian cancer development
PDF
Understanding DNA methylation and nucleosome organization in cancer cells using single molecule sequencing
PDF
The role of glucose-regulated proteins in endometrial and pancreatic cancers
PDF
DNA methylation changes in the development of lung adenocarcinoma
PDF
The role of PAX8 in epithelial ovarian carcinoma
PDF
Exploration of the roles of cancer stem cells and survivin in the pathogenesis and progression of prostate cancer
PDF
Genomic, transcriptomic and immunologic landscapes of human cancers
PDF
DNA methylation as a biomarker in human reproductive health and disease
PDF
Understanding the role of BRCA1 and Aurora A kinase in polyploidy development in ovarian carcinoma precursor cells
PDF
Identification of novel epigenetic biomarkers and microRNAs for cancer therapeutics
PDF
Investigating the role of SASH1 gene located on chromosome 6 in ovarian cancer
PDF
Identification of DNA methylation markers in diffuse large B-cell lymphoma
PDF
Functional characterization of colon cancer risk-associated enhancers: connecting risk loci to risk genes
PDF
Breast epithelial cell type specific enhancers and functional annotation of breast cancer risk loci
PDF
The effect of tumor-mediated immune suppression on prostate cancer immunotherapy
PDF
The cancer stem-like phenotype: therapeutics, phenotypic plasticity and mechanistic studies
Asset Metadata
Creator
Shen, Hui
(author)
Core Title
Integrative genomic and epigenomic analysis of human cancer
School
Keck School of Medicine
Degree
Doctor of Philosophy
Degree Program
Genetic, Molecular and Cellular Biology
Publication Date
05/13/2013
Defense Date
03/22/2013
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
cancer,clear cell renal cell carcinoma,epigenome,HNF1B,OAI-PMH Harvest,ovarian cancer
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Laird, Peter W. (
committee chair
), Berman, Benjamin P. (
committee member
), Pearce, Celeste Leigh (
committee member
)
Creator Email
shenhui1986@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-254284
Unique identifier
UC11295069
Identifier
etd-ShenHui-1682.pdf (filename),usctheses-c3-254284 (legacy record id)
Legacy Identifier
etd-ShenHui-1682.pdf
Dmrecord
254284
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Shen, Hui
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
clear cell renal cell carcinoma
epigenome
HNF1B
ovarian cancer