Array-Based Genomic and Epigenomic Studies in Healthy

Till mamma, pappa, bror och Lasse

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I Díaz de Ståhl TD, Sandgren J, Piotrowski A, Nord H, Anders- son R, Menzel U, Bogdan A, Thuresson AC, Poplawski A, von Tell D, Hansson CM, Elshafie AI, Elghazali G, Imreh S, Nor- denskjöld M, Upadhyaya M, Komorowski J, Bruder CE, Du- manski JP. Profiling of copy number variations (CNVs) in healthy individuals from three ethnic groups using a human genome 32 K BAC-clone-based array. Hum Mutat. 2008 Mar;29(3):398-408.

II Piotrowski A, Bruder CE, Andersson R, Díaz de Ståhl TD, Menzel U, Sandgren J, Poplawski A, von Tell D, Crasto C, Bogdan A Bartoszewski R, Bebok Z, Krzyzanowski M, Jan- kowski Z, Partridge EC, Komorowski J, Dumanski JP. Somatic mosaicism for copy number variation in differentiated human tissues. Hum Mutat. 2008 Sep;29(9):1118-24.

III Sandgren J, Díaz de Ståhl T, Andersson R, Menzel U, Pio- trowski U, Nord H, Bäckdahl M, Kiss N, Braukhoff M, Komo- rowski J, Dralle H, Hessman O, Larsson C, Åkerström G, Brud- er C, Dumanski J.P and Westin G. Recurrent genomic alterations in benign and malignant pheochromocytomas and paragangliomas revealed by whole-genome array comparative genomic hybridization analysis. Endocr Relat Cancer. 2010 June;17(3):561-79.

IV Sandgren J*, Andersson R*, Rada-Iglesias A, Enroth S, Du- manski J.P, Åkerström G, Komorowski J, Westin G and Wade- lius C. Integrative Epigenomic and Genomic Analysis of Ma- lignant Pheochromocytoma. *Both authors contributed equally. Exp Mol Med. 2010 July;42(7):484-502.

Reprints were made with permission from the respective publishers.

Related publications by the author

i Meyer-Rochow GY, Jackson NE, Conaglen JV, Whittle DE, Kunnimalaiyaan M, Chen H, Westin G, Sandgren J, Stalberg P, Khanafshar E, Shibru D, Duh QY, Clark OH, Kebebew E, Gill A, Clifton-Bligh R, Robinson BG, Benn DE, Sidhu SB. MicroRNA profiling of benign and malignant pheochromocytoma identifies novel diagnostic and therapeutic targets. En- docr Relat Cancer. 2010 Aug 16;17(3):835-846. ii Popławski AB, Jankowski M, Erickson SW, Díaz de Ståhl T, Partridge EC, Crasto C, Guo J, Gibson J, Menzel U, Bruder CE, Kaczmarczyk A, Benetkiewicz M, Andersson R, Sandgren J, Zegarska B, Bała D, Srutek E, Allison DB, Pio- trowski A, Zegarski W, Dumanski JP. Frequent genetic differences between matched primary and metastatic breast cancer provide an approach to identification of biomarkers for disease progression. Eur J Hum Genet. 2010 May;18(5):560- 8. iii Nord H, Segersten U, Sandgren J, Wester K, Busch C, Men- zel U, Komorowski J, Dumanski JP, Malmstrom PU, Díaz de Ståhl T. Focal amplifications are associated with high-grade and recurrences in stage Ta bladder carcinoma. Int J Cancer. 2010 Mar 15;126(6):1390-402. iv Nord H, Hartmann C, Andersson R, Menzel U, Pfeifer S, Piotrowski A, Bogdan A, Kloc W, Sandgren J, Olofsson T, Hesselager G, Blomquist E, Komorowski J, von Deimling A, Bruder CE, Dumanski JP, de Ståhl TD. Characterization of novel and complex genomic aberrations in glioblastoma using a 32K BAC array. Neuro Oncol. 2009 Dec;11(6):803-18.

v Bruder CE, Piotrowski A, Gijsbers AA, Andersson R, Erick- son S, Díaz de Ståhl T, Menzel U, Sandgren J, von Tell D, Poplawski A Crowley M, Crasto C, Partridge EC, Tiwari H, Allison DB, Komorowski J, van Ommen GJ, Boomsma DI, Pedersen NL, den Dunnen JT, Wirdefeldt K, Dumanski JP. Phenotypically concordant and discordant monozygotic twins display different DNA copy-number-variation profiles. Am J Hum Genet. 2008 Mar;82(3):763-71. vi Andersson R, Bruder CE, Piotrowski A, Menzel U, Nord H, Sandgren J, Hvidsten TR, Díaz de Ståhl T, Dumanski JP, Komorowski J. A segmental maximum a posteriori approach to genome-wide copy number profiling. Bioinformatics. 2008 Mar 15;24(6):751-8.

Contents

INTRODUCTION ...... 13 Human Genetic Variation ...... 15 Copy number variations ...... 16 Somatic mosaicism ...... 22 DNA copy number changes in the context of somatic mosaicism ...... 22 Epigenetics ...... 25 DNA Methylation ...... 25 Chromatin and genomic regulation ...... 26 Histone modifications ...... 27 Cancer ...... 30 Hallmarks of cancer ...... 30 Cancer genetics ...... 32 Genomic profiling of cancer ...... 33 Epigenetics in cancer ...... 34 Remarks on large scale cancer genetic/epigenetic research ...... 38 Pheochromocytoma ...... 40 Diagnosis and management of pheochromocytoma ...... 42 Malignant pheochromocytomas ...... 42 Familial pheochromocytomas ...... 45 Genetic testing in pheochromocytoma patients ...... 47 Signaling pathways and genes involved in pheochromocytoma development ...... 48 Somatic genetic and epigenetic alterations in pheochromocytomas .... 50 PRESENT INVESTIGATIONS ...... 53 Aims ...... 53 METHODS ...... 54 Array- CGH ...... 54 ChIP-chip ...... 56 SUMMARY OF INCLUDED PAPERS ...... 57 Paper I ...... 57 Profiling of Copy Number Variations (CNVs) in Healthy Individuals from Three Ethnic Groups Using a Human Genome 32 K BAC-Clone- Based Array ...... 57

Paper II ...... 58 Somatic Mosaicism for Copy Number Variation in Differentiated Human Tissue ...... 58 Paper III ...... 59 Recurrent Genomic Alterations in Benign and Malignant Pheochromocytomas and Paragangliomas Revealed by Whole-Genome Array-CGH Analysis ...... 59 Paper IV ...... 61 Integrative Epigenomic and Genomic Analysis of Malignant Pheochromocytoma ...... 61 CONCLUDING REMARKS ...... 63 Ongoing research and future perspectives ...... 63 General and future genomic and epigenomic views on pheochromocytoma and cancer ...... 66 ACKNOWLEDGMENTS ...... 69 REFERENCES ...... 73

Abbreviations

Array-CGH Array-based comparative genomic hybridization BAC Bacterial artificial chromosome ChIP Chromatin immunoprecipitation CNV Copy number variation DNA Deoxyribonucleic acid FoSteS Fork stalling and template switching GWAS Genome wide association studies H3K4me3 Histone 3 lysine 4 trimethylation H3K27me3 Histone 3 lysine 27 trimethylation kb kilo base pair LCR Low copy repeats LOH Loss of heterozygosity Mb Mega base pair MEN2 Multiple endocrine neoplasia type 2 miRNA microRNA MOR Minimal overlapping region MZ Monozygotic NAHR Nonallelic homologous recombination ncRNA non coding RNA NHEJ Nonhomologous end joining NF1 Neurofibromatosis type 1 PcG Polycomb group protein PCR Polymerase chain reaction PRC Polycomb repressive complex PGL Paraganglioma syndrome RNA Ribonucleic acid SD Segmental duplication SNP Single nucleotide polymorphism SNV Single nucleotide variant TSS Transcription start site VHL von Hippel-Lindau syndrome

INTRODUCTION

The genome can be described as the entirety of an organism’s hereditary information that lies in the DNA sequence. The human nuclear genome consists of more than 3 billion bases constituting the DNA molecules present in each cell of the body. The four different bases, adenine (A), thymidine (T), cytosine (C) and guanine (G) are assembled in a way that the DNA molecule forms a double-stranded antiparallel helix (1). The protein coding DNA sequences transcribed as mRNA are estimated to comprise 1-2% of the genome, recent findings though have shown that a much larger part of the genome is transcribed and many of these RNA molecules probably have regulatory functions (2). The DNA is well packed into chromosomes in each cell nuclei due to coiling of the DNA helix together with specialized proteins at several levels resulting in the chromatin (Figure 1). The chromatin is dynam- ically structured and can be re-organized in different contexts, apparently independent of the underlying DNA sequence. These changes of the chromatin can have great impact on heritable genome functions, such as gene activity. The term ‘epigenome’ was thus coined, where one genome can generate many 'epigenomes', manifested for instance during development and in different cell types as well as in pathogenic conditions, including cancer.

The accomplishment of sequencing the entire human genome in the beginning of the present century was a milestone in genetic research (3, 4). This effort, along with continuous methodological and technical advances which further enable the global study of also epigenomes, have paved the way for the plethora of important and exiting investigations addressing and revealing genetic and epigenetic impact in health and disease. The main interest of this thesis concern aspects of global genomic and epigenomic profiling in the context of normal and cancer cells.

Figure 1. Chromosomes and DNA. The nuclear genome is split into a set of linear DNA molecules, each contained in a chromosome. Chromosomes are much shorter than the DNA molecules that they contain: the average human chromosome has just under 5 cm of DNA. A highly organized packaging system is therefore needed to fit a DNA molecule into its chromosome. This is mainly dependent on the DNA associated histone proteins. When the nucleus divides, metaphase chromosomes form at a stage in the cell cycle after DNA replication has taken place so each nucleus con- tains two copies of its chromosomal DNA molecule (chromatide). The two copies are held together at the centromere. Telomeres are the terminal regions of the chromosomes. The human genome consists of 23 pair of chromosomes (T A Brown, Chapter 2.2, Genomes 2nd edition, Oxford: Wiley-Liss; 2002. Source of Figure: The Cancer Genome Atlas, http://cancergenome.nih.gov/media/chromosome.pdf).

14 Human Genetic Variation

The amount of knowledge regarding human genetic variation has expanded rapidly since the completion of the draft of the human genome sequence in 2001 (3, 4). Our modern understanding of human genetics embraces both the great diversity and complexity of the different forms of genetic variation that exist in the human genome, as well as the growing awareness that a significant portion of the genome is subjected to this diversity. Technical advances in the last few years have contributed greatly to the cataloguing and elucidation of the significance of DNA sequence variability in the human genome, and this area of research is continuously producing results. However, many questions remain to be answered, including the matter of the more precise location and frequencies of the variants as well as their implication on phenotypes in health and disease. The variation in the DNA sequence can involve from a single base pair (bp) up to several mega base pairs (Mb) and usually the variants are divided into two groups based on the nucleotide composition; single nucleotide variants (SNVs) and structural variants. In terms of frequency, the more common variants at a certain locus which are present in >1% of the alleles in the population, are called polymorphisms. Considerable effort has been focused on the discovery and understanding of the patterns of single nucleotide polymorphisms (SNPs) in the human genome. The estimated 17 million SNPs identified so far are deposited in the SNP database (dbSNP) (5). Novel SNVs are reported with each additional genome being sequenced, made possible due to the new high throughput technique of massive parallel sequencing (6-10). It is well known that numerous rare point mutations can cause sporadic disease and a fraction of SNPs have been shown to be linked to common human diseases in genome wide association studies (GWAS) (http://genome.gov/gwastudies).

Our knowledge about structural variants is however more recent and rudi- mentary compared to that of SNVs, due to later development of the methodological abilities in the detection of large rearrangements. Structural variants include insertions, deletions, block substitutions, duplications, inver- sions, and translocations of DNA sequences (11). An additional form of genetic variation is tandem repeats, often divided into short tandem repeats (STRs) and variable number tandem repeats (VNTRs) based on the sequence length. STRs and VNTRs are also known as micro and mini satellites, respectively. Although the DNA sequence in tandem repeats is simple compared to other types of more complex genetic variation, these sequences can be repeated from tens to hundreds of times, thereby creating genomic diversity (12, 13).

15 Copy number variations

In the preceding years, many genome wide studies have reported the presence of structural variation in the human genome (14-22). These studies have revealed that these events are widespread and common; even in healthy individuals and that they frequently involve gene rich regions. The majority of these findings concern copy number differences of DNA segments, known as copy number variations (CNVs) resulting from the extensive application of array comparative genomic hybridization (array-CHG). Array- CGH is a powerful method enabling the screeing of entire genomes for copy number imbalances at high resolution. The common definition of DNA copy number variation is as follows: a segment of DNA larger than 1kb displaying copy number imbalance between two genomes of the same species (23).

Two initial landmark studies reporting CNVs between normal human individuals were published in 2004 (14, 15). Since then many additional reports, using higher resolution genome analysis platforms, have advanced our knowledge of CNVs and a comprehensive genome wide map is beginning to emerge. A first generation human CNV map was constructed by the The Copy Number Variation Project which analysed 269 samples included in the HapMap project. This study revealed considerable variation between normal healthy individuals and more than 1447 CNV regions spanning 12% of the reference sequence were identified (17). A more recent report on the Hap- Map samples employing a hybrid genotyping array collecting both SNP and CNP data estimated however large scale CNVs to involve less than 5% of the human genome. This high resolution study (average limit of detection was 2kb) implied that previously reported CNVs regions seemed to be over- estimated in size (20).

Structural variants of 100 bp or larger are catalogued in the Database of Ge- nomic Variants (Figure 2). By March 2009, more than 38000 CNVs, and many more additional structural variants such as balanced rearrangements had been deposited in that database. The percentage of the reference genome covered by these CNVs is approximately 30%, which clearly outnumbers the fraction believed to be covered by SNPs, which is <1% (24). The limitations of the database however include the great variety of platforms used in the various studies. Thus, 30% may be an overestimation due to the inclusion of some low resolution studies but on the other hand, this figure may also represent an underestimation due to both the inability to detect small CNVs and the limitations of the current reference genome.

Figure 2. Current genome wide view of CNVs. Blue bars indicate reported CNVs; Red bars indicate reported inversion breakpoints; Green bars to the left indicate segmental duplications. Figure adopted from Database of Genomic Variants (DGV, http://projects.tcag.ca/variation), in may 2010. Original reference Iafrate AJ et al., Detection of large-scale variation in the human genome. Nature Genetics 2004 Sep;36(9):949-51).

In late 2009, a study was published utilizing an extremly high resolution array platform with the aim of generating a comprehensive map of relatively common CNVs. Nearly 12,000 CNV loci were identified in the screening of 40 individuals, and 8,599 of those were validated. Also, roughly 5,000 of the detected CNVs were genotyped in additional DNA samples derived from 450 Europeans, Africans or East Asians. The validated CNVs covered 3.7% of the genome and it was estimated that in average 0.78% of an individual’s genome consisted of CNV regions. Taking all samples in consideration, almost 40% of the CNVs overlapped with 13% of RefSeq genes. Gene ontology analysis showed enrichment of genes involved in extracellular biological processes such as cell adhesion, recognition, and communication while intracellular processes such as metabolic and biosynthetic pathways were

17 underrepresented in regions of CNV (21). It is now recognized that the difference in the genome sequence between any two individuals is significantly more extensive than earlier believed and also that CNVs contribute more to this variation than SNPs. SNPs are however the most abundant type of genetic variation in the human genome in terms of their number and they occur at an interval of about one SNP per every kilobase of DNA sequence throughout the genome. This is roughly equivalent to 3 million SNPs in each individual genome, representing 0.1% of the genome, and therefore the DNA sequence similarities of two individuals was previously estimated to be 99.9%. The more recent reports describing the thousands of CNVs that span hundreds of megabases, as well as emerging numerous indels and balanced rearrangements, have made it feasible to reject the earlier proposed high fraction of sequence similarity. Thus, the DNA sequences of individuals within and between populations are genetically more diverse and varied than previously thought (5). Furthermore, CNVs are also probably as important as SNPs in determining the differences between individuals (25). More recent results from the analysis of four fully sequenced human genomes, combined with targeted sequencing of structural variants greater than 8kb in length in eight human genomes have also provided insight about this topic. Based on this data it was estimated that it is expected that structural variants constitute between 9 and 25 Mb of the genome in an individual, corresponding to ~0.5 to 1%, underscoring the importance of these variants in genome evolution, health and disease (11).

The major fraction of CNVs is inherited, but variation may also arise de novo both in germline and in somatic cells (26). The mutation rate of point mutations is estimated to be approximately 2x10-8 per base per generation (27, 28). The mutation rate for CNV is more elusive, but several studies about de novo CNVs have estimated the range from 1.7x10-6 to 1.0x10-4 per locus per generation (29, 30) meaning a 100 to 10 000 times higher rate than the nucleotide substitution rate. Interestingly, while SNPs can mutate with roughly equal probability across the genome, the mutation rate for CNVs can vary widely depending on the genomic location. This is probably related to the underlying mechanisms causing the formation of CNVs, as well as to the regional chromatin architecture inciting genomic instability (24).

Mechanisms of CNV formation There are three primary mechanisms responsible for the generation of genomic rearrangements that are believed to account for the formation of the majority of CNVs in the human genome: Nonallelic homologous recombination (NAHR), nonhomologous end joining (NHEJ), and a third more recently proposed mechanism called fork stalling and template switching (FoSTeS).

18 NAHR Most recurrent genomic rearrangements are caused by NAHR between two low copy repeats (LCR), also known as segmental duplications (SD). LCR are region-specific DNA segments of approximate 10-300 kb in size and having > 95-97% similarity. Due to the high sequence identity, non-allelic copies of LCR can sometimes be aligned in meiosis and mitosis. This misa- lignment and the subsequent crossover between them results in genomic rearrangements in the progeny cells. When the two LCRs are located on the same chromosome and in direct orientation, NAHR between them causes duplication and/or deletion. In contrast, when LCRs lie on the same chromosome but in opposite orientation NAHR results in inversion of the fragment flanked by them. NAHR between repeats on different chromosomes can lead to translocation (31).

NHEJ In eukaryotic cells, NHEJ is a major mechanism of repair for double-strand breaks (DSBs). This process is also considered to be responsible for most of the rejoining of translocated chromosomes in cancer. NHEJ proceeds in four steps; detection of DSB, bridging of both broken DNA ends, modification of the ends to enable ligation, and finally ligation. Consequently, no sequence similarity is required for this mechanism to occur. Additionally, inclusion or exclusion of nucleotides can be observed at the rejoining site. It is proposed that NHEJ is stimulated and regulated by the genomic architecture to some extent, and breakpoints have often been observed in regions of repetitive elements (31).

FoSTeS FoSTes is a more recent DNA replication error mechanism model that potentially explains additional complex genomic rearrangements not formed by NAHR and NHEJ. FoSTeS occur when the DNA replication fork stalls at one position during replication and the lagging strand disengages from the original template, transfers and then anneals by virtue of micro homology at the 3´end, to another replication fork in physical proximity. The lagging strand then “primes” and restarts the DNA synthesis on the new fork. De- pending on the direction of the fork progression and whether the lagging or leading strand in the new fork was used as template and copied, the erro- neously incorporated fragment from the new fork could be in direct or in- verted orientation to its original location. Moreover, depending on whether the new fork is located downstream or upstream, the template switching results in either a deletion or a duplication. This whole procedure can occur multiple times in a series thus explaining regions of complex rearrangements with normal diploid DNA copy number between deleted/duplicated genomic segments. FoSTeS is probably also influenced by local genomic architecture

19 (micro homology) but unlike NAHR and NHEJ, this mechanism is based on the translocation of the end of a single nascent strand, so the genomic architecture facilitating FoSTeS may function without DSB (24, 31). It has also been proposed that FoSTeS can generate both large CNVs of several megabases (32, 33) as well as rearrange single exons (34). It is moreover believed to be a major mechanism behind the formation of non-recurrent CNVs and even LCRs, thereby providing for NAHR and creating a predisposition to additional genomic rearrangements in future generations (35).

Phenotypic consequences of CNVs The majority of common CNVs are most likely functionally neutral and do not result in a disease phenotype. There are however several ways by which CNVs can have a functional effect. For instance by affecting gene dosage, interrupting or fusing of genes, through position effects, and by unmasking recessive alleles or functional polymorphisms (24). CNVs involving dosage sensitive genes can alter gene expression levels and might thereby cause a clinical consequence (36). Affecting the gene dosage is believed to be one of the most common mechanisms for the functional consequences of CNV. However, a breakpoint for a specific CNV located within a functional gene may interrupt it and cause its inactivation and loss of function. Gene fusion can generate a gain of function mutation, a mechanisms prominent in certain cancers by somatic translocation events. Position effect means that a CNV can have an effect on the expression or regulation on a nearby gene by re- moving or altering a regulatory sequence (37). A deletion of one allele might unmask another recessive allele or functional polymorphism (38).

Copy number alterations involving large chromosomal regions have long been associated with human diseases. One of the best known is Down’s syndrome caused by trisomy of chromosome 21 (39). Subsequently, submicros- copic DNA copy number changes were also found to be associated to different pathological conditions and the reports of clinical findings relating to chromosomal imbalances are constantly accumulating. These so called genomic disorders are archived in the DECIPHER (Database of Chromosomal Imbalance and Phenotype in Human using Ensemble Resources) and in May 2009 over 2,200 cases of almost 50 diseases have been included in the database (24). Examples of genomic disorders linked to small DNA copy number changes are Charcot-Marie-Tooth disease type 1a (CMT1A) and hereditary neuropathy with liability to pressure palsies (HNPP), which both are due to recurrent alterations on a relatively limited region of genome. Copy number change of the dosage sensitive gene PMP22 on chromosome 17p12 are responsible for most of these cases, where duplication or reciprocal deletion causes CMT1A or HNPP respectively. However, the genomic rearrangements responsible for genomic disorders are generally not so small and often contain several genes. For instance a recurrent microduplication at chromo-

20 some band 22q11.2 which is associated with DiGeorge and velocardifacial syndromes (24).

Specific CNVs have also been directly linked to many common complex traits, including autism, schizophrenia, epilepsy, Parkinson disease and Alz- heimer’s disease (reviewed in 24). Relatively recently several genome-wide studies have described multiple disease-causing CNVs, including de novo events, in autism, schizophrenia, and also in mental retardation patients (23, 40). Most of the variants detected in these studies are rare however and only affects a fraction of the patients. A well known CNV-dependent susceptibility locus linked to a complex disease includes the gene CCL3L1 which varies in copy number between individuals and populations. In this case, the susceptibility concerns the risk of HIV infection and a low copy number of this gene encoding a chemokine receptor ligand is associated with higher infection vulnerability. Interestingly, copy number as high as ten have been observed in African populations (41, 42).

As mentioned above, so far SNPs have been the most studied form of genetic variation and GWAS have reported associations between numerous SNPs and common complex diseases as well as to various cancers. However, most of the variants identified confer relatively small increments in risk and explain only a small portion of the heritability seen in families with disease. This has raised the question of “missing heritability”. It is believed that CNVs and other genomic rearrangements may account for some of this un- explained heritability. Until now, CNVs with disease association seem to follow the same pattern as that of SNPs, in that rare variants tend to show a strong association effects whereas the common variants show a more modest effect, yet are carried by a large portion of the population (43). Additional large population studies including detection of also rare variants will probably provide more knowledge about the importance of CNVs in many complex traits. It is expected that the amount of data regarding CNVs will continue to grow with the application of higher resolution and wide-ranging assays, including the so called massive parallel sequencing techniques. This will certainly contribute to further understanding of genetic variation and its significance in health and disease.

21 Somatic mosaicism

It is generally assumed that the DNA sequences in the majority of the healthy cells of an organism are identical. An adult human is though composed of ~1014 cells so during the development and life span of an individual there are copius rounds of DNA replication and cell division, in which some errors could be introduced leading to somatic genetic change. Mosaicism is defined as the presence of somatic cells within a single organism that have different genetic composition, despite being derived from the same zygote. Mosaicism should not be confused with chimerism, which is also defined as the co-existence of genetically distinct cell lines in the same organism. How- ever in the case of chimerism, the origins of the distinct cell lines are from different individuals. Examples of chimerism are when cells persist in a re- cipient after a transplant and the persistence of fetal cells in the mother after the child is born (44). Most somatic changes that occur during development and life have no consequence for the phenotype. The outcome is however dependent on the specific DNA sequence being altered as well as the cell type and the developmental window during which the change/mutation occur. Cancer is obviously a prominent example of somatic alterations with deleterious effects for the affected individual (this will be addressed in more detail in separate chapters further below). Somatic mutations found in the context of mosaicism, including cancer, can range from point mutations and indels to large structural rearrangements encompassing thousands or millions of base pairs and even whole chromosomes (44).

The extent of somatic genetic variation and its functional consequences in disease but also in normal cells are not as well studied as the inter-individual genetic differences. An example of a well known physiological somatic mosaicism is however the somatic rearrangement of immunoglobulin and T-cell receptor (TCR) genes that takes place in B and T lymphocytes. These genes are inactive in most cells but undergo controlled genetic reorganization to become activated. This leads to production of monospecific antibodies or TCRs in B or T-lymphocytes, respectively, a diversity important for the immune response system (45).

DNA copy number changes in the context of somatic mosaicism Due to current microrarray based technical advances, as well increased variety and number of samples and tissues examined, our awareness of mosaic CNVs within an individual is improving. In comparison to karyotyping the resolution of array-CGH is much higher and consequently smaller alterations can be more easily detected. As mosaicism per definition affects a fraction of cells and array-CGH analysis routinely includes combined DNA from many

22 cells there is a limit of CNV identification concerning also possibly somatically developed events. Nevertheless, sensitivity is improving and somatic mosaicism has been detected as present in 5-20% of cells examined (46, 47).

Somatic mosaicism coupled to genes causing Mendelian disorders is reported more and more frequently. As a part of its nature mosaicism can result in a milder phenotype or unmask expression of a mutation thatotherwise would have been lethal for the embryo. It is therefore likely that many cases of somatic mosaicism are not discovered clinically, since the patients may show reduced pathogenic phenotype due to low numbers of cell carrying the mutation. The fraction of mutated cells will depend on the time point of the de novo change in the developing embryo, the later it happens the fewer number of cells become affected in the adult individual. Also, supporting the probable bias regarding the low detection of cases with mosaicism, could be that it mostly occurs in sporadic cases, with no family history of disease, and subsequently these individuals and/or their mutations are not identified to the same extent as in patients within families carrying known heritable syndromes. Some examples of diseases that can be caused by mosaic CNVs are Duchenne Muscular Dystrophy (DMD), which is a neuromuscular disorder due to mutations in the DMD gene on chromosome X, and Neurofibromato- sis Type 1 (NF1). DMD and NF1 cases also present with mutations in germline. NF1 is an autosomal dominant disorder which will be discussed briefly later, since a small fraction of these patients develops malignancies such as pheochromocytoma. The more common characteristics of NF1 include neurofibromas, café au lait spots, cognitive defects and lish nodules. The mutated NF1 gene is located on chromosome 17 and deletions occurring in this region are believed to be the product of NAHR (44, 48).

Meiosis seems to be very error prone and two thirds to three fourths of all conceptions are lost, of which half are due to chromosomal aneuploidy (48). It is well known that very few chromosomal aneuploidy events are tolerable during meiosis, no autosomal monosomies are compatible with life and only trisomy 13, 18 and 21 results in live birth. Sex chromosome aneuploidy is on the other hand more viable. Thus, it has been long known that several large chromosomal alterations and aneuploidies exist as somatic mosaicism in different syndromes, such as trisomy 8, trisomy 9, tetrasomy 12, monosomy 14 and trisomy 20 as well as the three aneuploidies mentioned above (48). There are also reports of aneuploidy and other large scale chromosomal alterations in the context of somatic mosaicism that does not give rise to pathogenic conditions in certain cases. This includes, trisomy 13, tetrasomy 5, trisomy and tetrasomy 9 as well as jumping Robertsonian Translocation resulting in mosaicism for rob(13;13) and rob(13;15) (49-52). All of which have been found in normal subjects. The application of BAC clone based array-CGH in clinic genetics laboratories has allowed for the detection of

23 also sub-chromosomal alterations and it has been estimated that 0.4% of all screened patients have a pathogenic somatic mosaicism utilizing this method (53, 54). More high resolution arrays have also recently been used and one study reported a frequency of 1% for aneuploidy mosaicism in patients referred to clinical diagnostic testing (47).

Finally, monozygotic (MZ) twins are an excellent resource to study somatic mosaicism, including genetic changes such as CNVs. Numerous comparative studies of monozygotic twins have been conducted in the attempts to elucidate the contribution of environmental versus genetic factors in a wide range of diseases and behavioral aspects. MZ twins are derived from a single fertilization and post-zygotily splits and develop into two embryos. MZ twins therefore share the same DNA and are supposed to be genetically identical. The phenotypical difference observed between a pair of MZ twins should then primarily be due to the environmental factors and differences in their epigenomes. However, a rather recent study concerning detection of CNVs has shown the existence of somatic mosaicism in MZ twins. In this study, 19 pairs of MZ twins were analyzed, which were then further divided into 2 groups where nine pairs were discordant for Parkinson´s disease or related disorder and ten pairs were concordant. CNV loci that differed between pair of twins were found in both groups. The results from this report suggest that genetic differences between MZ twins may be more common than earlier believed. This study also supported the idea to use discordant MZ twins in the search for novel disease associated de novo gene mutations (46). Importantly, as mentioned above, epigenetic changes such as DNA methylation may also explain phenotypic differences between MZ twins (55, 56).

Mosaicism clearly contributes more to human variation and disease than previously estimated. Additional studies in many more individuals, including MZ twins, and several tissues of different lineages and ages including cell sorting, as well as technical advances regarding resolution and sensitivity of detection will be required to shed more light on the frequency and importance of somatic genetic mosaicism.

24 Epigenetics

Multicellular organisms generally have nearly identical genomic sequence in all of the different cell types and tissues compromising the adult individual, even if intra-individual genetic differences exists as discussed above. Yet the gene expression pattern for each distinct cell type is unique (57, 58), a cell- specificity that cannot be explained by limited genomic regions of somatic mosaicism. The maintenance of this cellular identity is obviously important during development and must be retained during mitotic cell divisions. The explanation for these specific transcription programs lies within the term “epigenetics”. The name epigenetic is derived from the Greek “epi” meaning “upon” (59) and the term epigenetics was first used in 1939 by Conrad Wad- dington to describe “the causal interactions between genes and their prod- ucts, which bring the phenotype into being” (60). Later, Arthur Riggs and colleagues defined epigenetics as “the study of mitotically and/or meiotically heritable changes in gene function that cannot be explained by changes in DNA sequence” (61). Today the term has broadened to include both gene transcription changes that are heritable and transient in nature (62, 63). In other words, the epigenetic mechanisms are those which affect the structure and/or function of the chromatin in order to regulate gene expression, without alterations of the DNA sequence itself. This regulation is believed to occur by the ability of epigenetic signals to influence the accessibility of transcription factors to DNA, thereby resulting in activation or repression of gene transcription. Moreover, epigenetic modifications can also directly serve as docking sites for other protein complexes that in turn affect the chromatin structure and lead to functional consequences. Epigenetic mechanisms include DNA methylation, posttranslational histone modifications, histone variants, and small interfering RNAs.

DNA Methylation DNA methylation is the most well-studied and understood epigenetic modification and importantly, it is most often associated with gene silencing. Today strong evidence exist for DNA methylation as a epigenetic stable inheritance between cell divisions (64). DNA methylation has been shown to play a critical role in the control of various cellular processes such as embryonic development, transcription, X chromosome inactivation and genomic imprinting (65). DNA methylation patterns are established during embryonic development by the de novo DNA methyltransferases DNMT3A/3B and are maintained in the subsuquent cell divisions by the maintenance methyltransferase DNMT1 (66). A broad range of developmental abnormalities and diseases have been linked to the disruption of methylation patterns (65, 67). DNA methylation is also of great interest in cancer research and, as

25 discussed briefly below, seems to be of great significance in tumourgenesis. DNA methylation occurs at the C5 position of cytosines that precede gua- nines, so-called CpG dinucleotides (68-70). In total, 70-80% of the CpGs in the somatic human genome are methylated which constitutes approximately 1% of all bases (71). In normal cells DNA methylation is mainly found in repetitive regions, such as satellite repeats, long and short interspersed transposable elements (LINES and SINES respectively) and endogenous retrovi- ruses (72). CpGs are not randomly distributed, rather they are clustered together within the genome in certain CpG rich regions known as CpG islands. These islands span the 5´-end of the regulatory regions of approximately 60% of all genes (73). The majority of CpG islands remain unmethylated in normal cells, although particular promoter CpGs for tissue- and germline- specific genes are often methylated (74). More data from genome wide studies of the methylation status in human cells, termed epigenomic profiling, are emerging and this will provide many new insights as to the full significance of epigentics. For instance it was recently reported in a global survey that the most tissue-specific DNA methylation in brain, liver and spleen was not at the CpG islands but at regions of lower CpG density, located up to 2kb from the CpG island named island shores (75).

Chromatin and genomic regulation In epigenetics, DNA in the context of chromatin is the main consideration. In humans and other eukaryotes the nucleosome is the basal unit of the chromatin which in turn constitute two copies of each of the four core histones (H3, H4, H2A, and H2A). Approximately 146 bp of DNA is wrapped around these octamers of histone proteins (76, 77). Chromatin is however not only a packaging tool, it is dynamic and adjusts in response to regulatory signals needed to program appropriate cellular pathways. Histones are particularly important for such functions and these evolutionarily conserved proteins contain an accessible amino-terminal tail which can be subjected to posttranslational modifications (78). Chromatin is further characterized by histone variants and the spacing between nucleosomes (known as the nucleosome occupancy). In fact, alternative histones comprise another regulatory mechanism shown to be involved in gene regulation and often also in DNA damage (79). Histone variants are structurally distinct non-typical ver- sions of histone proteins encoded by independent genes and are often subjected to regulation that differs from that of the typical histones (80). There is also a growing understanding of the roles of RNA in gene regulation on a post transcriptional level, which represent an additional epigenetic mechanism. For example, small non-coding micro RNAs (miRNA) have been found to be involved in the regulation of a variety of biological functions such as developmental and metabolic processes but are also clearly linked to tumourigenesis (81). miRNA degrades mRNA or inhibits translation by sequence

26 specific base-pairing in the 3´untranslated regions of the target mRNAs (82). Moreover, other types of non-coding RNA (ncRNA) molecules have recently been shown to regulate chromatin modifying protein complexes (83, 84) and as the density of transcribed RNAs in cells seems to be much higher than previously appreciated (2) it is proposed that ncRNA –based mechanism s could account for greater regulatory control in health and disease than currently known today.

Histone modifications Histone modification, which have been shown to be heritable during cell divisions (85), is acknowledged as a key process in epigenetic regulation. The dynamic alterations of chromatin rely heavily on histone modifications and it is recognized that the core histones can be subjected to more than 100 different modifications including acetylation, methylation, phosphorylation and ubiqutination (86, 87) . There are primarily specific positions at the amino-terminal end of the histone tail that are modified and today most is known about lysine acetylation and methylation. Acetylation of lysine is associated with chromatin accessibility (euchromatin) and transcriptional activity, whereas lysine methylation can have different effects depending on which resi- due is modified; methylation of histone 3 lysine 4 (H3K4) and H3 lysine 36 (H3K36) is associated with transcribed chromatin. In contrast, methylation of H3 lysine 9 (H3K9) and H3 lysine 27 (H3K27) seem to lead to repression and a more compact chromatin structure (heterochromatin) (86) (Figure 3). Histones, in addition to having a significant role in the regulation of gene expression are also important in DNA damage repair, DNA replication and recombination (88). Due to the numerous potential modifications, a histone code hypothesis has been proposed and is defined as follows: distinct modifications, on one or more histone tails, act sequentially or in combination to form a ”histone code” that is read by other proteins to bring about distinct downstream events (89). Moreover, apart from the different possible covalent chemical modifications, the methylation of lysine can be either in a mono, di or tri state (me1, me2, me3). Taken together this implies that the potential number of combinations is enormous.

Figure 3. Cytosine and histone methylation. A. The repressive cytosine methylation (orange) and histone modification H3K27me3 (orange) as well as the active histone modification H3K4me3 (green) are illustrated. B. Diverse additional histone modifications (Figure adopted from Bernstein BE, Meissner A, Lander ES. The mammalian epigenome. Cell 2007;128(4):669-81.).

Epigenomic profiling The major methodological approach in the study of histone modifications is chromatin immunoprecipitation (ChIP). This technique utilizes a chromatin extract of the cells or tissue of interest and an antibody against the histone modification under investigation is used for immunoprecipitation of cross- linked DNA. By also carrying about several ChIPs from the same extracts using different antibodies conclusions can then be drawn regarding the combinations of histone modifications at the chromatin at that specific time point. In order to systematically investigate genome-wide histone modification patterns ChIP has been applied together with microarrays (ChIP-chip) or more recently with massive parallel sequencing (ChIP-seq). Such epigenomic studies have provided understanding of the global patterns of a number of histone modifications in various cell types (90-93) and also contributed to the understanding of their impact in disease including cancer (see below). With regards to the histone code of normal cells a comprehensive study in human T cells utilized ChIP-seq to map as many as 39 different modifications (93). In total they identified 13 main patterns present in at least 62 genes supporting studies in yeast and Drosophila that only a fraction of the possible myriad of combinations of histone modifications actually exists in vivo (94, 95). Notably, in the aforementioned T-cell study a backbone of 17 modifications was detected in roughly 30% of more than 12000 investigated promotors and importantly, most of these genes were shown to be highly expressed. Studies addressing the histone modification patterns in cell diffe-

28 rentiation and development have revealed that high levels of histone acetylation is important in the maintenance of an undifferentiated state in the stem cell (96, 97). Moreover, so called “bivalent” enrichment of the transcriptional repressive mark H3K27me3 and the activating mark H3K4me3 on many developmental regulatory genes have commonly been observed in mouse and human embryonic pluripotent stem cells (92, 98-100). During cell differentiation the bivalent mark is lost from the majority of genes and either one of the two marks often remains. The bivalent genes in general have low expression meaning that H3K27me3 is the dominant mark; H3K4me3 seems to keep genes poised for activation and does not necessarily mean transcription activity (101, 102). Trimethylation of H3K4 is one of the most studied lysine modification and it is found at the nucleosome that flanks the nucleosome- free region which coincides with the transcription start site (TSS) of actively transcribed genes in all eukaryotes examined to date (91, 102). These sites of H3K4me3 enrichment are also clearly linked to the location of the methyltransferase enzymes responsible for the modification (103, 104). The view is that H3K4me3 does not directly affect transcription but rather specifies recruitment of downstream effectors. Supporting this notion, factors associated with transcription have been reported to interact with H3K4me3 (105, 106). Several studies have recently shown that novel promotors can be identified on the basis of the H3K4me3 chromatin signature including TSSs for ncRNAs (91, 107).

Methylation of the histone modifications H3K4me3 and H3K27me3 Transcriptional cofactors containing a so called SET domain predominantly perform histone methylation. Two different groups of protein complexes conveying this enzymatic activity have been identified which also show an association with either transcription maintenance or repression. These two complexes are termed Trithorax and Polycomb group proteins. Several trithorax proteins are H3K4 methyltransferases, for instance MLL and SMYD3. Polycomb group proteins (PcG) such as EZH2 have H3K27 methyltransferase activity. Moreover, the PcG proteins form complexes called Polycomb repressive complexes (PRCs) and the vital role of these in development and somatic cell differentiation due to their effect on chromatin are more and more recognized (108). The PRC1 complex inhibits gene activity by interfering with the movement of the nucleosome and by blocking recruitment of transcriptional activation factors (109). EZH2 is a member of the PRC2 complex and the trimethylation of H3K27 is a prerequisite for the action of PRC1 (109). Given the critical role of H3K27me3 in the balance of gene activity is it not surprising to find alterations of this system in diseases such as cancer and will be discussed in more detail below.

29 Cancer

Cancer is a global health problem responsible for one in eight deaths worldwide (110). Of all reported deaths in Sweden, neoplasms are the second most common cause; just after diseases of the circulatory system (Socialstyrel- sen). Cancer is characterized by the relatively uncontrolled proliferation of cells which also can progress and invade neighbouring tissues and metastasize to distant organs. Most deaths are due to metastatic spread, so prevention of metastatic tumours is of high clinical value and attention (111). Despite its common properties cancer is a complex collection of diseases that can arise from most of the cell types and organs of the body. There are more than 100 distinct cancer types with diverse risk factors and epidemiology (112). Genetic, and/or epigenetic alterations are believed to be the cause of cancer. Such alterations can be inherited in the germline or acquired somatically. A variety of DNA damage mechanisms exist which cause the cancer genotype, as discussed below, nevertheless, tumour causing alterations may also be the consequence of the effects of carcinogens, such as tobacco smoke, radiation, chemicals, or infectious agents. The majority of all human cancers are known to occur as sporadic cases with no family history of disease. Cancer is thought to develop through a multistep process of an accumulation of the plethora of possible mutations, consistent with Darwinian evolution principles. In this way random mutations and epigenetic changes reactivate or modify signaling pathways, a process which is followed by the clonal selec- tion of cells that can survive and proliferate under circumstances that normally would be lethal (113, 114).

Hallmarks of cancer The classical phenotypical characteristics that have been described as the hallmarks of cancer include i) unlimited proliferation potential, ii) self- sufficiency in growth signals, iii) resistance to antiproliferative and iv) apoptotic cues, v) attraction of new blood vessels to bring oxygen and nutrients, vi) evasion of immune detection and ultimately vii) metastasize to distant organs (115, 116). These acquired properties are mainly due to the irreversi- ble alterations present in the cancer genome, namely consisting of single bases (point mutation), insertion and deletions of small DNA segments, rearrangements including translocations of DNA sequences and copy number imbalances of large genomic regions. The cancer genome is also subjected to potentially reversible epigenetic changes involving DNA methylation and histone modification status, as well as regulatory RNAs (Figure 4). In addition, exogenous sources of new DNA sequences from viruses such as papil- loma virus which can integrate into the genome of cancer cells can also lead to tumour development (117).

Recently several additional hallmarks of cancer has been proposed, based more on the common phenotypes observed than on the initiating tumouri- genic events. These are collectively called the stress phenotypes of cancer and include: viii) DNA damage/replication stress, ix) proteotoxic stress, x) mitotic stress, xi) metabolic stress and xii) oxidative stress (113). A brief description of these important tumour characteristics follows here, starting with viii) DNA damage. Most solid tumours are subjected to genomic instability resulting in the generation of point mutations, deletions, complex chromosomal rearrangements and extensive aneuploidy (118). The DNA damage seen in early stage tumours are believed to be caused by several mechanisms, for instance the shortening of telomeres due to replication in the absence of sufficient telomerase activity leading to double-strand breaks (DSBs) at the telomeric ends. The subsequent fusion of these unprotected ends leads to breakage-fusion-bridge cycles that result in translocations and gene amplifications (119). Mutations in genes involved in DNA repair programs such as excision, crosslink and mismatch repair, and/or the DNA damage stress response (DDR) pathways (120) such as ATM and p53, can also lead to increased DNA damage and inappropriate cell-cycle progression and genomic instability (121). Cancer cells, as opposed to normal cells, overcome the antiproliferative effects of DNA damage and obviously continue to replicate. ix) Proteotoxic stress is described as an increased amount of toxic unfolded proteins which aggregate in the cell and thereby putting stress on the folding and degradation machinery of the cell (122).

x) A mitotic stress phenomenon is chromosomal instability (CIN) which is manifested as an increased rate of gains and losses of whole chromosomes due to mitotic mis-segration (123). The CIN phenotype can result from defects in a variety of pathways involved in mitosis, recently it has been shown that the presence of an extra centrosome in cancer cells likely contributes to chromosome mis-segregation during cell division (124). In hereditary cancer, the presence of genomic instability has been linked to mutations of DNA repair genes, however, this association is not as clear in many sporadic cases (125). xi) Metabolic stress refers to the observation that cancer cells, again in opposition to normal cells, produce the majority of their energy through glycolysis, a phenomen known as the Warburg effect (126). Normally, cells generate the energy resource molecule ATP via the mitochondrial oxidative phosphorylation mechanism. This alteration is advantageous for tumour cells which can then adapt to low oxygen environments more easily as well as the side effect of producing lactic acid, which promotes tumour invasion and suppresses immune surveillance (113). xii) Finally, oxidative stress which relates to the increased presence of reactive oxygen species (ROS) possibly in part due to down-regulation of mitochondrial function (127). ROS are yet another likely to contributor to DNA damage in cancer cells (113).

31 Cancer genetics The first findings pointing to the central role of the genome in cancer development were made in the late nineteenth and the early twentieth centuries by David von Hanseman and Theodir Bovier (128, 129). By investigating divid- ing cancer cells under the microscope they observed bizarre chromosomal alterations which led to the theory that cancer originates from and is characterized by abnormal clones of cells carrying severe aberrations in the hereditary material. Subsequent crucial findings in the field of cancer genetics followed; the discovery of the DNA molecule as the inheritance material (130), its structure (1) and the observation that DNA-damaging agents causing DNA mutation generated cancer (131). Additional more sophisticated chromosomal analyses showed that specific and recurrent genomic alterations were associated with certain types of cancer, such as the translocation between chromosome 9 and 22 in chronic myeloid leukemia (“Philadelphia chromosome”) (132). In 1982, the first isolated single DNA sequence change able or sufficient to cause cancer was reported. The mutation is a G > T substitution in the HRAS gene that lead to an amino acid switch from gly- cine to valine and result in an activated Ras protein (133). The search for genetic background to cancer development and progression has been massive and continues today, with the expectation of even more exciting results especially with the application of new technological and methodological advances.

Somatic mutations accumulate over time and can be categorized according to their role in cancer development as driver or passenger mutations. Driver mutations promote a growth advantage for the cell and are positively selected during cancer evolution. Passenger mutations, on the other hand, do not contribute directly to cancer development, but are carried along during clonal expansion. Approximately 420 driver genes have been identified today (http://www.sanger.ac.uk/genetics/CGP/Census/). These cancer genes have been reported to show recurrent somatic mutations in tumours with evidence of contribution to the disorder. Mutations in about 10% of these genes are also found in germline, predisposing to an increased risk of developing cancer. Moreover, approximately 90% of the cancer genes acts domi- nantly, i.e. the mutation in one allele is sufficient to contribute to cancer development. In the case of dominant acting cancer genes (also known as oncogenes), the somatic mutation is constrained. Missense mutations, in- frame insertions and deletions, and gene amplification are common events. The most frequent event is however a result of genomic rearrangements, such as the joining of sequence from two different genes to create a fusion gene or that the cancer gene is positioned behind an active promoter. Both scenarios give rise to aberrant upregulation of gene expression. The remaining 10% of cancer genes acts in a recessive manner; these genes are often

32 called tumour suppressor genes. The patterns of mutations differ between the dominant and recessive cancer genes. Tumour suppressor genes can be inactivated by a wide range of mutations, such as single nucleotide substitution to whole gene deletion as well as epigenetic silencing, which may result in complete loss of protein function (112). According to the two-hit mutational model by Knudson, two genetic events (acting on each allele) are required for a tumour suppressor gene to lose its function (134). However, haploinsufficiency may also contribute to tumourigenesis, in a dominant acting effect manner.

Genomic profiling of cancer During the past decade the understanding of the cancer genome landscape has greatly improved due to the advancement of genomic profiling techniques, from karyotyping via array-CGH to the currently increasingly used massive parallel sequencing. The improved resolution has been described as more than a million fold (135). Several hundreds of array-CGH studies with variable resolution have so far been performed on various cancer samples and numerous tumour-specific chromosomal aberrations have been identified, including important findings of candidate genes within those genomic regions. DNA copy number profiling has also aided in classifying subgroups of tumours that correlates to clinical outcome (135, 136).

The genomic instability of cancer cells, manifested and profiled as somatic DNA copy number changes in array-CGH analysis (Figure 4), is one major responsible cause of the observed altered gene expression known to contribute to tumour development (135, 137-140). Amplification of genomic segments may lead to increased mRNA and protein expression and several oncogenes are known to reside in those regions of multiple DNA copies (140, 141). Likewise, deletion of DNA segments may lead to haploinsufficiency or the complete loss of expression of tumour suppressor genes, with fatal consequences. The implementation of also array-based gene expression analysis has revealed novel transcriptional signatures correlating with clinical outcome (142, 143). Moreover, integration of genome wide expression profiling with copy number analysis have significantly increased the power to identify candidate cancer genes (144, 145). Emphasizing the importance of detecting recurrent somatic copy number changes in cancer and in an approach to catalogue those that are associated with the development of many major types of cancer a comprehensive array-based study was recently published analyzing more 3000 tumour samples of 26 different tissues. Due to the large size of the common aberrations involving whole chromosome arms they focused on short (focal) alterations (median length of 1.8Mb) and indentified 158 frequent (1.5%-14%) and independent regions across all samples. The most frequent of these focal amplified and deleted regions contained the MYC and

33 CDKN1A/B gene, respectively. Interestingly, 122 of the 158 copy number changed segments did not contain a known cancer gene. Moreover, nearly 200 focal regions were identified in individual cancer types enabling for detection of candidate genes. Conclusively, different tumours may harbor similar genomic alterations that are of importance, but clearly great diversity exists and more studies are needed to elucidate the functional relevant events (146).

Epigenetics in cancer

Cancer and DNA methylation The importance of epigenomic changes in tumour biology is gradually becoming obvious. Loss of DNA methylation at CpGs was the first epigenetic change to be found in cancer cells (69, 70). At present, this epigenetic mechanism is the most studied in cancer, and probably also plays a major part in tumourgenesis. The pattern of DNA methylation in cancer cells is greatly altered and is largely manifested as wide spread hypomethylation. This occurs mainly in regions of repetitive DNA and gene-poor areas but also in coding and intronic regions. DNA methylation is believed to be an early event and subsequently during tumourigenesis the degree of global hypomethylation increases (147). Different explanations have been proposed to describe the contribution of hypomethylation to cancer progression such as increase in genomic instability, reactivation of transposable elements, and activation of proliferation prone genes and loss of imprinting (82). Loss of DNA methylation can lead to mitotic recombination, resulting in deletions, translocations and chromosomal instability. Genomic imprinting, the silencing of one of the two parental alleles, is partly maintained by differential methylation of regions near or within imprinted genes, a process normally programmed in the germline. Disruption of imprinting results in an unbalanced abundance of the imprinted gene product. For instance, the loss of imprinting of the IGF2 gene increases IGF2 protein levels which are associated with an elevated risk of developing a wide number of cancers such as liver, lung and colon (65, 148, 149). Moreover, genomes of cancer cells are also characteristically found to contain de novo hypermethylation at specific regions, preferentially in CpG islands of tumour suppressor genes and mi- croRNA genes. Inactivation of tumour suppressor genes due to DNA methylation is a major cause of cancer. For instance, the aberrant de novo hypermethylation of CpG islands in promotors of known tumour suppressor genes such as CDKN2A/p16INK4a, MLH1, VHL and E-Cadherin has been shown to correlate with decreased expression in several human cancers (150).

Further high resolution studies on the global changes in DNA methylation of individual samples will be presented in the coming years revealing addition-

34 al tumour type-specific methylomes. Certain CpG island hypermethylation profiles have already been shown to differ between cancer forms (151, 152). The increased recognition of such epigenomic patterns occurring both in sporadic and inherited cancers could certainly have implications for therapy strategies. DNA hypermethylation markers are also being developed for complementary diagnostic and prognostic tools as well as predictors of response to treatment (82).

Histone methylation in cancer It is continuously emerging that besides DNA methylation the epigenetic regulation of histone modifications plays a major role in the control of transcription, cell differentation as well as in the development of cancer (87, 153). Histone methylation was described over 40 years ago but the actual enzymes responsible for the methylation were reported much later, starting in the mid 1990s. Many of these enzymes have also subsequently been frequently observed to be differentially expressed in tumours and/or involved in oncogenic translocation events. For instance the proteins MLL1, SMYD3, EZH2 and DOT1L involved in methylating histone residues such as H3K4, H3K27 and H3K79 have all been associated with cancer development (154). This also applies to certain proteins which are responsible for the demethy- lating activity of H3K4me3, H3K27me3 and H3K9me3 namely LSD1, UTX, PLU-1 and JMJD2C (155-158). Chromosomal translocations of the MLL1 gene are for example found in over 70% of infant leukemia cases and these aberrations correlate with poor outcome (159). Another H3K4 methyltransferase, SMYD3, has been observed to be upregulated in prostate cancer cells (160) and in colorectal, hepatic and breast cancer (161) indicative of H3K4 hypermethylation association and cancer progression.

Moreover, the H3K27 trimethylase EZH2 has been reported as overexpressed in numerous tumour tissues such as breast, bladder, colon and gastric cancers, which is believed to lead to the functional consequence of silencing of genes that promote differentiation and inhibits proliferation (tumour suppressor genes) (162). Studies have also shown that H3K27me3 can target loci for de novo DNA methylation in cancer cells and the connection between these two different actions of gene silencing is an active area of research (163, 164) . DNMTs as well as histone deacetylases (HDACs), have been reported to be recruited to CpG regions by the PRC2 complex and contribute to the silencing of the nearby gene (163). Also, H3K27me3 was shown to be a premark in colon cancer for de novo CpG methylation (165). However, DNA methylation independent changes of histone modification pattern have also been directly linked to cancer development and specifical- ly, inactivation of genes marked by H3K27me3 in the absence of DNA methylation has been reported (166). Notably, several histone acetyltransferase

35 (HAT) genes are altered in cancer resulting in loss of the active H3 acetylation mark at tumour suppressor genes (167).

Genome wide investigations of histone modifications in cancer genomes The advent of the array technology has in recent years led to a number of genome wide ChIP-chip studies investigating the histone modification status also in tumour cells. Given the importance of these chromatin marks it is believed that specific modifications can distinguish different cancers and possibly be used in prognosis as well as in the identification of targets for drug development. Global patterns of acetylation and methylation of histones H3 and H4 are associated with multiple cancer types and, in prostate cancer for instance, modification patterns distinguish disease subtypes and can predict patient outcome (162, 168-170). In one of these studies it was shown that prostate cancer patients displaying decreased levels of H3K4me2 together with H3K18Ac had poorer prognosis than those patients with higher levels of these histone modifications (169). Moreover, a general reduction of H4K16ac and H4K20me3 is seen during tumour development in a wide range of cancers (168) and alterations in H3K9 methylation patterns have also been associated with aberrant gene silencing in cancer (171). Interes- tingly, H3K27me3 also shows the potential to be used as a prognostic marker. Genes marked with H3K27me3 in metastatic prostate cancer cells corres- ponded to a high degree with PcG silenced genes in stem cells, a result that was used to develop a PcG repression signature which predicted clinical outcome. The signature also proved valid for breast cancer (170). Additional recent high resolution whole genome studies in prostate cancer cells have also shown the distribution of H3K4me3 and H3K27me3 and their correlation to gene expression as well as to upregulation of the methylating enzymes responsible for these modifications (160).

Figure 4. The main types of genomic and epigenomic aberrations are illustrated together with examples of how they can be detected. a, Changes in DNA sequence, such as point mutations, can be assessed by DNA-sequencing techniques. b, Changes in genomic organization can be assessed by using fluorescence in situ hybridization. In the example shown, DNA segments are exchanged between the two (blue and green) DNA molecules. c, Changes in DNA copy number, such as those that result from amplification, can be assessed by using comparative genomic hybridization. d, Changes in DNA methylation and the resultant changes in chromatin structure can be assessed by using chromatin immunoprecipitation plus microarray analysis of immunoprecipitated DNA. Each of these types of change can alter the expression levels of genes or microRNAs (referred to here as genetic elements of interest, GEOIs), alter the splicing patterns of transcripts, or change gene function through mutation or through creating chimaeric genes. Many of these events can be as assessed by microarray analysis. These changes may ultimately translate into altered functions, leading to a diseased state, such as cancer (Figure adopted from Chin L, Gray JW. Translating insights from the cancer genome into clinical practice. Nature 2008;452(7187):553-63.).

37 Remarks on large scale cancer genetic/epigenetic research Many important findings regarding cancer genetics have been reported with the use of methods such as microarrays which enable the identification of amplified and deleted genes, along with the re-sequencing of selected regions of cancer genomes post PCR-amplification. New high through put technical advances with massive parallel sequencing will however contribute enormously to the knowledge regarding mutations in cancer cells. Recently, data from such studies have become available revealing the complexity of the cancer genome. Even though several well known oncogenes and tumour suppressor genes such as PI3K, Ras, p53, PTEN, Rb, and CDKN2A/p16INK4a are frequently found mutated in cancer, it is becoming clear that apparently there are a high number of low frequency changes which are important in a subset of tumours. The tumour genomes and exomes sequenced so far have shown an extraordinary diversity of mutations (172-177) and each tumour sample harbours a complex combination of rare mutations thought to drive the cancer phenotype. But interestingly, despite the tumour heterogeneity and the fact that cancer genomes contain many infrequently and a few frequently mutated genes, signaling pathway analysis have been shown to significantly reduce the complexity (178). A limited number of affected shared pathways were discerned in a major fraction of tumours in a study of exonic mutations in pancreas cancer for instance (173). The large scale surveys of various tumour genomes have identified new possible cancer genes that may constitute potential therapeutic targets (157, 179-185). Functional studies are needed to discern driver genes from passenger genes (186).

The continuous decoding of individual cancer genomes will hopefully lead to more accurate diagnosis and prognosis and also guide personalized treatment. The insight into frequent cancer mutations has led to development of new drugs, such as the antibody trastuzumab for HER2 positive breast cancer and imitanib targeting the BCR-ABL tyrosine kinase in chronic myeloid leukemia (187, 188). More targeted therapies are to come, even if the recent advances in dicerning the diversity and complexity of mutations in individual tumours may suggest it to be also prudent to consider the development of therapeutics that target functional pathways or processes rather than individual genes and/or proteins that have been shown to be relatively commonly mutated.

The recent and ongoing advances in the field of epigentics and cancer have led to the realization that the epigenome may be as important as the genome itself when it comes to regulation of essential cellular processes and also in the cause and progression of cancer. A deeper understanding of the different global chromatin regulators and their distribution holds promises for an improved understanding of cancer cells, and will provide discovery of diagnos-

38 tic markers and development in specific therapeutic strategies that may aid in the treatment of cancer patients.

In the past few years, several initiatives have begun to search for somatic alterations that underlie human cancer more systematically and comprehen- sively. In 2008, a massive effort was launched by the then formed Interna- tional Cancer Genome Consortium (ICGC). The aim is to coordinate large- scale cancer genomic studies in tumours from 50 different types including the ones with highest incidence and mortality worldwide. More than 25 000 samples will be investigated at the genomic, epigenomic and transcriptome level with high resolution and coverage sequencing methods. The combination of these sophisticated and comprehensive analyses with correlation to clinical data hold promise for revealing the repertoire of disease causing mutations, the defining tumour subtypes for prognosis and diagnosis, as well as enabling the development of new cancer therapies (185).

39 Pheochromocytoma

Pheochromocytomas are rare neuroendocrine tumours of the sympathetic nervous tissue. The term pheochromocytoma relates to the color the tumour cells acquire when stained with chromium salts. In Greek phios means dusky, chroma means color and cytoma means tumour. Pheochromocytoma was first described in 1886 and the first surgical resection of a pheochromocytoma was performed in Lausanne, Switzerland in 1926 by Roux, later the same year Charles Mayo made the first surgical removal in USA (189). Pheochromocytomas are most often (80-90%) found in the adrenal medulla where they arise from catecholamine producing chromaffin cells, derived from the embryonic neural crest (Figure 5) (190). The adrenal glands sit on top of the kidneys and each gland consists of two distinct layers, the adrenal cortex and the medulla, which both produce hormones. The adrenal cortex is devoted to the synthesis of corticosteroid hormones from cholesterol and this tissue layer surrounds the catecholamine secreting medulla situated as the core of the gland.

Approximately 10-20% of the pheochromocytomas are located outside the adrenals, often referred to as extra-adrenal pheochromocytoma, or sympathetic paraganglioma (190). Paraganglioma however refers to all neoplasias of the paraganglion nervous system, consisting of both sympathetic and parasympathetic paraganglia located along the large vessels from the base of the skull to the pelvis. Paraganglioma of the parasympathetic nervous system are mostly situated in the head and neck region, and rarely secrete significant amounts of catecholamines (191). The sympathetic paraganglioma, are most common in the upper abdomen, but may occur in Zuckerkandl´s paraganglia at the aortic bifurcation, in the middle mediastinum and involve the heart, and in the wall of the urinary bladder (190).

Figure 5. A, Anatomical distributions of chromaffin tissue (Figure adopted from Lack E. Tumours of the adrenal gland and extra-adrenal paraganglia. In Atlas of Tumors Pathologhy. Armed Forces Institute of Pathology: Washington, DC, 1997, 261-267). B,A surgically removed pheochromocytoma. Photo provided by Professor Göran Åkerström, Uppsala University Hospital.

The clinical manifestation of pheochromocytoma results from excessive catecholamine secretion. The continuously or intermittently overproduced catecholamines released to the blood stream are epinephrine, norepinephrine and more rarely dopamine. Sympathetic paragangliomas however rarely secret any ephinephrine at all (191). The three hormones mentioned are derived from the amino acid tyrosine, with dopamine as the precursor for norepinephrine while epinephrine is synthesized due to additional modifications of norepinephrine. Epinephrine and norepinephrine are well known to sti- mulate alpha and beta adrenergic receptors causing a sympathetic “fight-or- flight” response. The biological effects include elevated blood pressure, increased heart rate and blood glucose levels. The most common symptoms of patients with pheochromocytoma are subsequently episodic headache, sweating and palpitation due the excessive stimuli of this system. Persistent hypertension is also frequently considered part of the clinical presentation

41 (192). The overall estimated prevalence of pheochromocytomas and paragangliomas is 1 or fewer cases per 100 000 individuals (193-195). About 15% of patients are asymptomatic and the diagnosis is often made in the context of diagnostic procedures for other purposes, such tumours are named incidentalomas (196).

Diagnosis and management of pheochromocytoma Pheochromocytomas are diagnosed by sensitive biochemical tests and then confirmed by specific imaging studies. The initial assessment of this catecholamine producing tumour is most often to measure in plasma or urine for metanephrines, which is a metabolite of epinephrine. Computed tomography (CT) and magnetic resonance imaging (MRI) are commonly used methods to localize tumours following positive biochemical testing. Pheochromocytoma cells express specific catecholamine plasma membrane and vesicular transport system enabling for functional imaging in patients either with a pheochromcytoma confirmed biochemically but when CT and MRI fail to show a tumour or when there is a suspicion of malignancy and metastatic spread. Iodine-131-labled metaiodobenzylguanoidine ([131I]MIBG) is a radiolabeled molecule similar to norepinephrine often used for this purpose since it loca- lizes to chromaffin cells. Another more recent important imaging technique is positron emission tomography (PET). Several different agents are being used and studies have shown benefits, especially for localization of small tumours throughout the body (192). After precise localization the tumour is removed by surgery most often with a laparoscopic approach. After the removal of the tumours the prospect, in the majority of cases with benign pheochromocytomas, is that all symptoms will be relieved and that the patient will be cured. Appropriate hypertensive drugs are used to manage hypertension, prevent cardio vascular complications and diminish the magni- tude of postoperative hypotension. This is because during the operation, massive catecholamine release may occur which effects have to be blocked (191, 192).

Malignant pheochromocytomas The majority of pheochromocytomas and abdominal paragangliomas are benign. Malignant pheochromocytomas account for approximately 10% of all cases although are more common among abdominal paragangliomas, where they constitute between 15-35% of all tumours (197, 198). Or if related to mutation in certain susceptibility SDH-gene, even higher frequencies are observed, as discussed below (199, 200). The World Health Organization definition of malignant pheochromocytoma from 2004 is the presence of metastases (201) but this distinction between benign and malignant pheochromocytomas does not account for recurrent or locally invasive tumours

42 which are also acknowledged as malignant criteria’s. As a matter of fact, currently these manifestations especially the evidence of metastasis which is most widely accepted, are the only means of diagnosing malignancy. Distant metastasis mainly spread to the liver, lymph nodes, lung and/or bones and can appear as long as 20 years after the initial presentation of the primary tumour (202). The current lack of appropriate clinical, biochemical, histopathological or genetic markers/predictors of malignancy make life-long follow up of patients with chromaffin cell tumours mandatory (197). However, some suggestions such as the combined use of the proliferative marker Ki-67 and the increased gene expression of human telomerase transcriptase (hTERT), may have diagnostic value of malignancy. Moreover, large tumour size and weight, low expression of the neuropeptide Y (NPY) and increased expression of the vascular endothelial growth factor (VEGF), heat shock protein -90 (HSP-90) as well as the endotehlin receptor type A and B have also been linked to malignant pheochromocytomas (reviewed in 197). Mar- kedly raised levels of plasma chromogranin A (CgA), a protein stored and released along with catecholamines, has also been claimed to indicate malignancy (197). However, these histopathological features, mainly detected by immunohistochemical methods, along with the observed increased mRNA and protein expression levels, are not widely accepted and none is specific for the disease. A histological scaling system exists, named pheochromocytoma of the adrenal gland scaled score (PASS) which uses a range of criteria to group adrenal pheochromocytomas with potentially aggressive behavior and those likely to behave in a benign way. However, this sceme cannot properly assign all the malignant tumours in the potential aggressive groups and significant interobserver and intraobserver variation occurs in the as- signment of PASS, thus further validation and refinement is needed, if possible (192, 203). The need for prognostic biomarkers of malignancy remains as a crucially important issue for pheochromocytoma.

In recent years a number of expression microarray studies have been performed addressing the issue of the specific transcriptome profiles in malignant pheochromocytomas. Several candidate genes with differential expression between malignant and benign cases have been proposed as a result. The concordance between the different studies is however not easily seen. The results therefore need to be validated in larger cohorts of clinical samples, although it still demonstrates the potential of the approach of genetic profiling in the discrimination of malignant pheochromocytomas (204-208). Also recently, additional candidate genes have been associated to malignant pheochromocytoma using methods allowing investigations of a limited number of genes. One interesting candidate is the known oncogene ERBB-2 located on chromosome 17, which displayed both increased protein expression and DNA copy number in malignant compared to benign cases (209). Other studies investigating genes involved in cell motility and invasive be-

43 havior showed an overrepresentation of high expression of the genes Snail and Twist in malignant pheochromocytomas (210, 211). Moreover, down regulation of 6 tumour suppressor genes (nm23-H1, TIMP-4, BRMS-1, TXNIP, CRSP-3, E-Cad) has been reported more often in malignant cases compared to benign cases (212). The well-known tumour suppressor gene, PTEN, located on chromosome 10, has also been investigated. Even if the mutation analysis in malignant and benign pheochromocytomas showed negative results in all included tumours in one report, LOH analysis conducted in the same study demonstrated loss of the PTEN loci in a higher fraction in the malignant cases compared to the benign ones (213). A conditional Pten knock-out mouse that develops chromaffin tissue derived lung metastasis has further shown its use as a model for metastatic pheochromocytoma since the genomic aberrations detected with array-CGH correlate to findings in human pheochromcytomas (214).

In humans, certain chromosomal aberrations have been more frequently observed in malignant cases compared to benign pheochromocytomas. These are important observations that may guide the discovery of genes that are located in these regions and their significance in tumour progression. How- ever, the rather few CGH reports on this topic point to somewhat different chromosomal regions with malignant association. The major finding in one CGH survey of pheochromocytoma and abdominal paraganglioma suggested that alterations on chromosome 11 is of importance for distinguishing between benign and malignant cases (215) and also that gain of the region 11cen-q13 was more common in abdominal paragangliomas. This was confirmed in another report with a set of both pheochromocytomas and head and neck paragangliomas were it was shown that this region were overrepre- sented in malignant cases of both tumour groups (216). Loss of chromosome arm 6q has also been associated to malignant Pheochromocytoma (Figure 6) (217). Gain of chromosome arms 17q, 12q and 20q as well as loss if chromosome arm 9p were observed in a higher frequency in metastazing pheochromocytomas compared to non-metastazing cases in an additional study, gain at 17q was however the most significant finding (218).

The overall survival rate for a patient with metastatic disease is 50% at 5 years. Morbidity and mortality are often related to tumour burden because of high circulation of catecholamines and the mass effect of metastases (192). The results of treatment of malignant disease are generally poor, since the tumours often are resistant to chemotherapy and radiotherapy. The management of malignant pheochromocytoma includes pharmacological control of symptoms and tumour mass reduction followed by radionuclide treatment and/or chemotherapy. Reduction of tumour mass is achieved by surgical resection of the primary tumour and local or distant metastases (197). For target radiotherapy the 131I-MIBG is used ant its labeling with radioactive

44 isotopes makes it useful both for diagnostic and therapeutic purposes (195). However, the chemotherapy and radiotherapy treatments now used for dec- ades, even if modified over the years, do not target specific pathways involved in tumour development but instead exhibit general cell toxicity with limited therapeutic benefit. With the increasing data on altered pathways and genes in pheochromocytoma will possibly allow more targeted therapies with hopefully improved benefits for the patients.

Familial pheochromocytomas The majority of pheochromocytomas and paragangliomas occur sporadically and prior to the year 2000 it was generally accepted that 10% of cases were associated with familial syndromes. However, it is now recognized that the fraction of germline mutations in the genes causing the hereditary disorders is 12-16% in apparent sporadic cases (219-221) implying that roughly 25% of all cases may be related to the hereditary syndromes (199). Germline mutations in six genes have been associated with familial pheochromocytomas; the von Hippel-Lindau gene (VHL) causing the von Hippel-Lindau (VHL) syndrome, the REarrenged during Transfection gene (RET), leading to multiple endocrine neoplasia type 2 (MEN2), the neurofibromatosis type 1 gene (NF1), associated with Neurofibromatosis type 1 (NF1) disease, and the genes encoding subunits B, D and C of mitochondrial succinate dehydrogenase (SDHB, SDHD and SDHC), which are associated with familial paraganglioma/pheochromocytoma syndromes (222-228). Recently it has also been shown that familial paraganglioma cases exists with germline mutations in the genes SDH5 (SDHAF2) and SDHA (229-231). Truly sporadic pheochromocytomas however rarely present with somatic mutations in any of the genes responsible for the hereditary syndromes (232-235).

VHL The VHL syndrome is an autosomal dominant disease characterized by the development of highly vascularized tumours of mesenchymal and neural crest derived organs. The incidence of VHL is one in 36 000 live births and the disease generally becomes evident between the ages of 20 and 40 (236). The VHL gene is localized on chromosome 3p25-26 and is a tumour suppressor. The encoded protein (pVHL) is involved in blood vessel formation by regulating the activity of the transcription factor hypoxia inducible factor 1α (HIF-1α). pVHL inhibits the accumulation of hypoxia induced proteins through the ubiquitin-mediated degradation of HIF-1α, under normoxemic conditions. Mutations of VHL lead to that the regulation of genes such as the VEGF and other genes involved in cellular growth seems to be lost as a consequence of the reduced HIF degradation, predisposing the VHL carrier to both benign and malignant tumours in multiple organs. VHL disease is divided into four subtypes (1, 2A, B and C) based on its clinical expression,

45 and genotype-phenotype correlations exist. Patients with VHL type 1 have loss of pVHL function due to gene deletions or specific missense mutations and do not develop pheochromocytoma but instead retinal or central nervous system hemangioblastoma and/or renal carcinoma. Patients with type 2 disease have mainly specific missense mutations and develop pheochromocytomas (roughly 20% of patients), hemangioblastoma and are either at low (type 2A) or high (type 2B) risk for devolpment of renal cell carcinoma. A small percentage of patients with VHL has only pheochromocytoma (type 2C) (191). VHL type 1 is believed to be due to loss of pVHL function whereas type 2C is thought to be caused by gain of pVHL function and the co- occurrence of loss and gain of protein function is thought to result in type 2A and 2B (237). Approximately 3-11% of patients with apparently sporadic pheochromocytomas will have a VHL germline mutation (192). Malignancy in VHL patients is reported in less than 5% of patients (190).

MEN2 MEN2 is a rare autosomal dominant tumour syndrome with an estimated prevalence of 2-5 per 100 000 in the general population. There are two forms of MEN2; MEN2A and 2B, where the former is characterized by medullary thyroid carcinoma (MTC) in all patients, pheochromocytoma in 50% and primary hyperparathyroidism in 20-30%. MEN2B patients also suffer from MTC and pheochromocytomas as well as mucosal neuromas on the lips and tongue, ganglioneuromatosis in the gastrointestinal tracts and skeletal abnormalities (190). MEN2 often manifests relatively early in life, between 5 and 20 years of age. Activating germline mutations of the oncogene RET on chromosome 10q11 are responsible for the inheritance of MEN2 (223, 238). The RET gene encodes a transmembrane receptor tyrosine kinase involved in the regulation of cell proliferation and apoptosis. Its´ expression is restricted to neural crest derived tissues (191). Different mutations in the RET gene with a clear genotype-phenotype correlation exist. Mutations in codon 634 are highly frequent and 85% of patients with MEN2A present with these mutations. Roughly 95% of the MEN2B patients present with point mutations in codon 918 (190). Mutations in those codons are also highly associated with pheochromocytoma development. Somatic mutations of RET have also been detected in 15% and 3% of benign and malignant sporadic pheochromocytoma, respectively. However, interestingly it has been reported an overexpression of the RET gene in the majority of (50-70%) of sporadic pheochromocytomas that do not harbor any mutations, amplifications or rearrangements of RET. This indicates that RET could play a significant role in sporadic pheochromocytoma development, but does not seem to promote malignant behavior (239).

46 NF1 Pheochromocytoma constitutes a minor, but established feature of NF1, an autosomal dominant disorder with an incidence of 1 in 3000 individuals (190). The estimated frequency of pheochromocytoma is 0.1-5.7% but rises to 20-50% among hypertensive patients with NF1. The diagnosis of NF1 is usually made by the following criteria: greater than six café au lait spots, more than two neurofibromas and axially freckling. The disease is caused by inactivating mutations of neurofibromin 1, a tumour suppressor gene, which encodes a GTPase-activating protein involved in the inhibition of Ras activity, which controls cellular growth and differentiation (191). The NF1 gene is located on chromosome 17q11.2 and it is one of the largest known genes. NF1 consists of 60 exons spanning more than 350kb of DNA sequence, hampering screening for mutations (190).

Paraganglioma/Pheochromocytoma syndromes Germline mutations in the genes SDHB, SDHC and SDHD encoding succinate dehydrogenases subunits result in familial paraganglioma syndromes (PGL). SDHB mutations appear to be most common, with an overall frequency of 1.7-6.7% in patients presenting with pheochromocytoma (240). PGL-SDHB is an autosomal dominant syndrome characterized by sympathetic paragangliomas, parasymaptetic paragangliomas and pheochromocytomas. The syndrome is caused by inactivation mutations of the tumour suppressor gene SDHB on chromosome band 1p35-36. Mutations of this gene cause mitochrondrial complex II destabilization and may activate the hypoxic/angiogenic pathway. Importantly, malignant pheochromocytomas are reported in such a high frequency as 35% (241) in these patients and for sympathetic paragangliomas up to 50% of tumours are malignant (200, 242). Mutation in the SDHC gene located at 1q21 represents an additional PGL syndrome characterized of benign head and neck paragangliomas. No mutations have been found in pheochromocytoma patients (190).The PGL-SDHD syndrome is due to mutations in SDHD gene located on chromosome 11q21- 23. Patients usually develop head and neck parasympathetic paragangliomas and occasionally pheochromocytomas (243). The hereditary nature of the disease may be masked by maternal imprinting of SDHD, resulting in that tumours only develops if mutation in the gene is inherited from the paternal side (244).

Genetic testing in pheochromocytoma patients The recent studies showing higher prevalence of germline mutions particularly in the RET, VHL, SDHB and SDHB genes than earlier believed also in non-syndromic patients, have raised the question of the need of genetic testing in an increased number of patients suffering from pheochromocytoma,

47 with or without a positive family history. Because most inherited pheochromocytoma present before the age of 50, it has been suggested that all patients younger than 50 years old should be referred for genetic counseling. Also, patients with extra-adrenal, bilateral or multiple pheochromocytomas should always be referred for genetic assessment regardless of age. Such manifestations are frequent in the context of the heritable syndromes. Specif- ic testing algorithms have been developed taking into account clinical characteristics and prevalence of mutations in the different syndrome associated genes, supposed to help limit the cost of testing and direct to which gene/genes to be tested first (190).

Signaling pathways and genes involved in pheochromocytoma development The genes responsible for the development of pheochromocytoma in the above mentioned familial syndromes may seem to affect different pathways and have unique functions according to their known physiological role. Yet histopathologically and clinically indistinguishable tumours arise in individuals carrying these various mutations, as well as sporadic tumours with no mutations in the known familial genes. A report on a large set of different syndromic and sporadic pheochromocytomas/paragangliomas addressing this genetic uniformity has contributed to the unraveling this apparent heterogeneity. In this study Dahia et al described two groups of different tumours with similarities in their transcription profiles (245). It was shown that VHL-, SDHB-, and SDHD mutant tumours as well as a subgroup of sporadic tumours comprised one group, while tumours with RET or NF1 mutations as well as another subgroup of sporadic cases gave rise to a second separate group, or cluster, with distinct expression pattern. This finding of two distinct clusters have shed light on what different major signaling pathways are disrupted during tumourgenesis and in addition provided evidence of interaction between pathways affected by different known mutations in the same cluster.

The cluster associated with VHL, SDHB, or SDHD mutations showed a transcription signature characterized by genes related to hypoxia, extracellular matrix component and angiogenesis. This is also consistent with the role of pVHL in the regulation of hypoxia response. Increased levels of HIF-1 alter the regulation of various cell functions including glycolysis, apoptosis, cell survival, cell differentiation and angiogenesis (246). Besides its connection with pVHL, HIF-1 is also associated with the enzyme SDH. SDH proteins are necessary for mitochondrial oxiditave phosphorylation and are involved in the oxidation of succinate to fumarate in Krebs cycle. SDHB and SDHD mutations lead to accumulation of the substrate succinate, which can inhibit

48 HIF-1 prolyl hydroxylases, enzymes that targets HIF-1 for degradation. The subsequently increased HIF-1 activity and of hypoxic signals in these tumours lead to overexpression of angiogeneic and cell growth factors (195). In hypoxic conditions, or pseudohypoxic conditions, as in the tumours har- boring these mutations, the pVHL can not bind to HIF-1alpha and subsequently the ubiqutin mediated degradation fails. HIF-1 related pathways seem to play a significant role not only in the development of pheochromocytoma, but also in the progression to malignancy since SDHB mutant tumours are prone to behave in that way. The role of HIF-1 as a classical oncogene is however disputed since overexpression in some contexts seems to suppress instead of promote tumour growth (247). In contrast, the RET and NF1 tumour cluster was characterized by an increased expression of genes involved in RNA synthesis and metabolism. Mutations of the oncogene RET causes ligand independent activation of this tyrosine kinase which results in activation of multiple downstream transduction pathways involved in cell growth, survival and differentiation. The NF1 protein shows a region of similarity with the family of GTPase activation proteins which inhibit Ras activity. NF1 loss/mutations lead to structural abnormalities in this region resulting in reduced GTPase activity, which will in turn increase GTP- mediated activation of the RAS/RAF/MAPK pathway (247). In conclusion, both mutated RET and NF1 share the ability to activate these oncogenic signaling pathways. Interestingly, recent biochemical evidence suggests that RET, NF1, SDHD, SDHB, SDHC and VHL mutations converge into a common signaling pathway that disrupts the prolyl hydroxylases Egln3/PHD3- mediated apoptotic response of sympathetic neurons to limiting amounts of nerve growth factor (NGF) during development (248). This model suggests that cells that escape this physiologically induced death during the particular developmental window, later will give rise to pheochromocytomas or paragangliomas.

Recently two additional candidate genes have been proposed as predisposing for pheochromocytoma, namely KIF1Bβ and TMEM127 (249, 250). KIF1Bβ encodes kinesin that functions in axonal transport and was identified in a screen for downstream mediators of proapoptotic effect of the prolyl hydroxylase EGln3 in neuronal precursor cells. Mutant KIF1Bβ led to reduced Egln3 mediated apoptosis, consistent with a tumour suppressor function. KIF1Bβ maps to 1p36, a region often deleted in pheochromocytoma, however most tumours did not show any mutations on the remaining allele as would have been expected in a classical two-hit mode of tumour suppressor inactivation. A haploinsufficiency mechanism may however explain the tumourgeneic effect of KIF1Bβ. Transcriptional analysis of two tumours with mutated KIF1Bβ suggests similarities with mutant RET and NF1 expression profiles (247). Germline mutations in the TMEM127 gene have been found in a few familial and sporadic pheochromocytomas (250). Clus-

49 ter analysis of global gene expression confirmed that these tumours were associated to the NF1 and RET mutant signature. TMEM127, located on chromosome 2q11, is suggested to act as a tumour suppressor gene by limiting the mTOR signaling pathway.

Somatic genetic and epigenetic alterations in pheochromocytomas Several LOH and metaphase CGH studies have been conducted including also apparently sporadic pheochromocytomas and paragangliomas with the aim to identify chromosomal regions of importance for tumour development. More recently also the application of high resolution array-CGH has been used to survey individual chromosomes or the entire genome of these tumours for candidate genes with altered DNA copy number. From all these studies it has become evident that the most frequent DNA copy number imbalances in pheochromocytomas are loss of chromosome arms 1p, 3p/q, 11p/q, 17p and 22q. Gain of chromosomal material has been less frequent, and the most commonly affected chromosomal regions include 1q, 12q, 17q, 19p/q and 20q (Figure 6) (215-218, 236, 251-253). Particularly, deletions of chromosome arm 1p are highly frequent in pheochromocytomas and paragangliomas, indicative of an early event in tumourgenesis as well as for the location of candidate tumours suppressor gene/s on this autosome. A whole genome array-CGH study of sporadic pheochromocytomas published only last year observed a pattern of two subgroups of genomic alterations, with similarities to the changes seen with metaphase-CGH in MEN2 and VHL related tumours respectively (253). Van Neederveen et al. were able to distinguish between tumours with a high frequency of 1p loss with or without 3q loss which resembles the alterations seen in the majority of MEN2 cases and a different subgroup with 3p loss with or without 11p loss which is more often seen in VHL-related cases (239, 253). Even though several candidate regions have been proposed in these studies above, probably due to low resolution and/or the limited number of samples few narrow regions were presented encompassing obvious candidate genes that have been further confirmed to be of significance in pheochromocytoma development and progression. Notably, loss of chromosome arm 17p is a frequent event, but mutations of the well known tumour suppressor gene p53 at 17p13.1 are rarely seen (254). A recent mouse model however points to the importance of p53 in pheochromocytoma development, but only when this occurs in a background of predisposing mutations of the gene Rb1 (255). Rb1 has previously been reported as being down regulated in pheochromocytomas (239).

Figure 6. DNA copy number changes in 29 apparently sporadic pheochromcytomas revealed by metaphase-CGH by Dannenberg et al 2000. The vertical green lines on the right side of the chromosome ideograms indicate gains, the red on the left losses of the corresponding chromosome region (figure adopted from Dannenberg H, Speel EJ, Zhao J, Saremaslani P, van Der Harst E, Roth J, Heitz PU, Bonjer HJ, Dinjens WN, Mooi WJ, Komminoth P, de Krijger RR. Losses of chromosomes 1p and 3q are early genetic events in the development of sporadic pheochromocytomas. Am J Pathol 2000;157(2):353-9.).

Regarding global gene expression studies in pheochromocytoma, the majority of them has addressed and revealed the transcriptome differences between benign and malignant tumours as well as between the different familial and sporadic forms (204, 206, 207, 245, 256). Reports of up and down regulated genes in pheochromocytoma compared to normal adrenal medulla are mostly restricted to investigations of individual genes. For instance AKT, IGF2 and IL-13Rα2 have all been observed as overexpressed in the tumours (257-259). Occasionally gene expression has been studied also in combination with the DNA methylation status of the promotor region, as for the tumour suppressor gene CDKN2A/p16INK4a that has been shown to be methylated and down regulated both in pheochromocytomas and paragangliomas (260, 261). The RIZ1 gene has also been seen down regulated, as well as deleted at the frequently deleted 1p 36.21 region. However no DNA methylation was observed in the promoter region, indicative of other mechanisms of transcriptional regulation (262). Additional examples of hypermethylated

51 genes in pheochromocytoma include HIC1, CASP8, TNFRSF10A, RASSF1A, MSH2 and HSP47 (263-267). The majority of investigated genes has been observed with a higher frequency of promoter methylation in the familial tumours compared to the studied sporadic cases.

This together with the lack of genome wide high resolutions studies of epigenetic alterations in benign and malignant pheochromocytomas indicate that more epigenomic changes perhaps specific to the sporadic cases, and maybe also to progression to malignancy, are to be found. This also applies to the field of pheochromcytoma genomics, where future investigations hopefully will reveal novel mutations involved in the development for these catecholamine producing tumours.

52 PRESENT INVESTIGATIONS

I will here present the aims, the main methods and the summaries of the four publications on which the thesis is based. This will be followed by a brief discussion on the obtained results and the ongoing research as well as remarks in general regarding future research addressing genetic variation and cancer genetics/epigenetics.

Aims

Establish, validate, and apply a high resolution genomic microarray in the profiling of a large set of phenotypically normal individuals to further survey the extent of CNVs in the human genome (Paper I).

Explore the presence of somatic mosaicism of CNVs in healthy differentiated human tissues (Paper II).

Investigate chromosomal alterations in benign and malignant pheochromocytomas and paragangliomas to detect common regions of loss and gain that may be of importance for tumour development and malignancy (Paper III).

Apply a combinatorial methodological and analytic approach including genome-wide profiling of the histone modifications H3K4me3 and H3K27me3, DNA copy number and gene expression in a malignant pheochromocytoma sample for the identification of candidate genes (Paper IV).

53 METHODS

Array- CGH

Array-CGH is a well acknowledged method for the detection of DNA copy number imbalances. Extensive applications of array-CGH includes CNV identification in healthy and disease related genomes, as well in profiling tumour genomes often characterized by gross chromosomal alterations and/or high DNA copy number gains and losses.

DNA copy number alterations assessed by conventional cytogenetic banding techniques such as fluorescence in situ hybridization (FISH) allows under ideal conditions detection and visualization of chromosomal rearrangements of at least 2-10Mb in size. The sensitivity of FISH is limited both by the size of the probes used and by the fact that only the regions assessed by the probes can be investigated for DNA alterations. The development of comparative genomic hybridization (CGH) technique overcomes some of these limitations since it allows for whole genome detection of DNA copy number changes. CGH was first described for the analysis of solid tumours (268). The principle of the method is that copy number differences are judged by hybridizing together the differently labeled DNAs from a test (ex a tumour) and a reference (ex a healthy individual) to metaphase chromosome spreads (so called methapase-CGH). The ratios of the test versus reference signals are used to determine the relative copy number of the DNA sequence in the test genome in comparison to the reference genome. Metaphase CGH is ro- bust for large-scale alterations but the resolution of detection is poor, events less than 20Mb in size are difficult to detect. More recent is the array-based CGH technique (269, 270). In this method DNA probes representing artificial chromosomes segments are arrayed on a chip and CGH is used to test for increased or decreased dosage of the chromosomal regions of interest (Figure 7). In one single hybridization array-CGH can detect DNA copy number alterations in the whole genome if the DNA probes arrayed allow this coverage. It is today commonly used in analysis of chromosomal balances in tumours, mental retardation and other constitutional abnormalities, as well in studies of genetic variation in healthy individuals. The detection limits of array-CGH in terms of DNA copy number imbalances identified are dependent on the probe density and the resolution of the platform used. The

54 DNA probes ranges from genomic clones, most often BAC clone (150- 200kb in size), to oligonucleotides (25-85bp in size) (271) . There have been many developments in the array technology in the last years with the aim to increase the detection resolution of the platforms. This is obtained by both shorten and increase the number of hybridization targets spotted on the chip.

In the studies included in this thesis we used a genomic microarray consisting of 32 396 tiling BACs (32K BAC array) covering 99% of the human genome (272). This 32K BAC array was established, validated and applied by our research group and average resolution is less than 100Kb (Paper I). Normalization and analysis of the signals obtained from each BAC after hybridization and scanning were executed in the web based Data Warehouse maintained by the Linnaeus centre of bioinformatics (LCB) (LCB-DWH, http://dw.lcb.uu.se). The analysis of the experiments also included use of visualization and statistical tools. All these post-hybridization processing and analyzing applications were developed in collaboration with LCB, including an algorithm for copy number calling (273).

Figure 7. Array-CGH. The method is based on a two-color, competitive fluorescent in situ hybridization of differently labeled test and reference DNA on an array of genomic clones. If the clones represent the entire human genome, this will allow for genome-wide detection of deletions and gains/amplifications in a single hybridization.

55 ChIP-chip

The interaction between DNA and proteins is fundamental to biological processes and plays a major role in regulating gene expression and control- ling the availability of DNA for transcription, replication, and other functions. These interactions can be studied in a focused manner using the chromatin immunoprecipitation (ChIP) technique (274). ChIP is a powerful tool for the identification of proteins, including transcriptions factors and histones, which are associated to certain regions in chromatin, by using specific antibodies recognizing the protein of interest. It is well established that the recruitment of transcription factors correlates with observed differential gene expression, and more recently it has been shown that covalent modifications of histones in the promoter regions also contribute significantly to gene transcription regulation (275) . While these recruitment processes are dynamic events, ChIP assays provide a way to freeze the process chemically and examine the current state in the cells.

Generally, ChIP-chip assays can be done on cultured cells or tissue samples. The initial step in the basic protocol is to covalently crosslink DNA to the associated proteins by treating the cells with formaldehyde. The second step is to lyse the cells and isolate the chromatin and thirdly, fragment the DNA to desired size by sonication. In the fourth step the DNA fragments are immunoprecipitated using the specific antibody. The immunoprecipitated DNA-protein complex are in the fifth step de-cross-linked with low pH buf- fer and high temperature. And lastly the ChIP-DNA, as well as the input DNA used as genome reference DNA is purified by protease digestion. After PCR amplification, the DNA is labeled and hybridized to microarrays (therefore named ChIP-chip) for global investigation of histone enrichment regions (275, 276).

In the present investigation, Affymetrix high-density oligonucleotide tiling arrays with a 35 bp resolution of the whole repeat-free genome have been used. After scanning the arrays for probe signals, the analysis of ChIP-chip data includes identification of positive probes and ChIP-enriched binding regions and the mapping of those enrichments to the genes and/or genomic regions.

56 SUMMARY OF INCLUDED PAPERS

Paper I

Profiling of Copy Number Variations (CNVs) in Healthy Individuals from Three Ethnic Groups Using a Human Genome 32 K BAC-Clone-Based Array The most studied genetic variation so far has been single nucleotide polymorphisms, but more recently much attention has been focused on structural variations as technology development has allowed for the detection of these common types of genomic variants. CNV constitutes one major class of structural variation and increased knowledge regarding the occurrence of CNVs is expected to result in better understanding of different phenotypes including disease susceptibility and also contribute to insight about the genetic diversity and evolution. The main method enabling the identification of these unbalanced chromosomal changes is array-CGH.

In the present study we established a high resolution genomic microarray comprised of 32 396 BAC clones. This platform was carefully validated and subsequently used to profile a large cohort of healthy individuals, of three different ethnicities, for common and novel CNVs. All samples were hybridized against blood-derived DNA from the same healthy reference female of European ancestry. The establishment of the microarray included PCR amplification of the 32K BAC library and in-house printing. The validation experiments consisted of self-self hybridization of the female reference DNA with no aberrant BACs as well as characterization of CNVs in her genome by hybridization experiments against pools of other subjects. Additional validation experiments included analysis of previous genetically well characterized cell lines and tumour samples.

A total of 71 cases (44 men and 27 women) of three different ethnicities (33 Europeans, 24 Africans and 14 Asians respectively) were analyzed for CNVs using the array. The results from the genomic profiling revealed 1078 autosomal CNVs that involved at least two neighbouring BACs. The average number of CNVs in each individual was 15.1 and the average size was 358 Kbp. The sizes of deletions and gains were similar but a substantially larger

57 number of gains (835 vs 243) were detected. Clustering of individual overlapping CNVs to CNV regions (CNVRs) resulted in 315 autosomal CNVRs encompassing 118,1MB, which corresponds to ~3.5% of the genome. Alto- gether 62.5% of the identified CNVRs mapped to regions already annotated in the Database of Genomic Variants, indicative of the reliability of our findings. Consequently 37.5% represented new variants. As much as 87% of the 315 CNVRs harbored known genes. Moreover, 91 CNVRs contained at least 30% of segmental duplications (SD), reinforcing the notion of SDs as hot spots for chromosomal rearrangements. Unsupervised hierarchical clustering did not reveal any ethnic-specific clustering of the samples. We also detected that multiple regions at chromosome X in males displayed copy number ratio that would be indicative of gain but probably these are due to specific homology blocks to chromosome Y and thus not true CNVs. This is supported by the observation that none of these regions were found to vary in hybridi- zations involving two females. Many of these regions have however been reported by others as CNVs.

In conclusion, this report presents a reliable analysis platform for genome- wide high resolution CNV detection and contributes to the growing knowledge of human genetic variation.

Paper II

Somatic Mosaicism for Copy Number Variation in Differentiated Human Tissue Somatic mosaicism is defined by the presence of genetically distinct populations of somatic cells in a single organism. In pathogenic conditions somatic changes affecting known disease genes or large chromosomal regions are common. It has moreover generally been assumed that all the normal cells in an individual are genetically identical; however, research addressing the issue of somatic mosaicism in apparently healthy individuals is as yet not widely explored.

In this study we tested the hypothesis that stochastic subchromosomal CNVs are present in DNA from fully differentiated nomal tissues. Three male subjects were included, from which we extracted DNA from 11 to 12 tissues each. The 32K BAC array was used as the primary platform in determining CNVs of size 50kb or larger. Since the highest amount of DNA was obtained from the cortex of cerebellum this was used as the reference tissue against the other tissues of the same individual.

Our results showed that at least six loci presented with intra-individual imbalances involving one or more tissues in the three subjects. Five of these loci were composed of two or more overlapping BACs. The six CNV regions detected were distributed on five chromosomes and ranged in size from 81.7kb to 176kb. Three of these regions contained annotated genes and it is therefore probable that these CNVs could have phenotypical effects due to changes in gene expression. The CNV regions displaying somatic variation in this study are known to vary between unrelated individuals according to data submitted to Database of Genomic Variants. This suggests that some CNVs reported as germline may instead represent somatic events, since CNV studies generally only examine one tissue and perform no analysis of parental DNA which could verify if CNVs are inherited. Moreover, five of the six discovered somatic CNVs flanked gaps in the present genomic sequence. This indicates that the variability in these loci with regards to segmental duplications and repeat units, might also be a possible reason for the failure in deciphering the sequence in these parts of the genome.

In summary, the results from this genomic profiling, on a limited number of subjects, indeed showed that somatic mosiacim is relatively common in normal human cells. This finding calls for more extensive investigations of somatic genetic structural variation in a larger set of samples, possibly including microdissected tissues as well as sample collection at different time points throughout the life of the same individual in order to also examine the de novo CNV rate.

Paper III

Recurrent Genomic Alterations in Benign and Malignant Pheochromocytomas and Paragangliomas Revealed by Whole- Genome Array-CGH Analysis In this study we applied high resolution array-CGH for genomic profiling of a relatively large series of benign and malignant pheochromocytomas and paragangliomas (n= 53). Matched constitutional DNAs were also included enabling the identification of tumour-specific aberrations. Previously CGH and LOH reports have shown frequent deletions on chromosome arms 1p, 3q, 11p and 22q in pheochromocytoma. Here we intended to extend the findings in these rare catecholamine producing tumours which most often occur as benign and sporadic (see above). We also aimed to identify regions associated with malignancy.

The global DNA copy number analysis revealed losses and gains of whole chromosomes as well as more restricted regions on most autosomes in the studied set of tumours. Some chromosomal regions were more frequently affected and the overall most common events were losses on chromosome arm 1p. Interestingly, tumours that did not show 1p loss frequently demonstrated aberrations on chromosome 11 instead. In general, loss of genomic segments occurred more often than gains. However, gains were significantly more frequent in the group of malignant tumours. This group also presented a higher number of large alterations per sample (a mean of 7.2 versus 4.9 in benign tumours). We further observed an overrepresentation of gain on 19q, trisomy 12 and loss on 11q in malignant compared to benign pheochromocytomas, whereas gain on 1q was more commonly observed in the malignant paragangliomas in comparison to malignant pheochromocytomas.

Minimal overlapping regions (MORs) of aberrations were identified on 16 chromosomes in all pheochromocytomas (n=44). This included four frequent (61-75%) and narrow (3.29- 15.05Mb) MORs of deletion on chromosome arm 1p. Each of these encompassed candidate tumour suppressor genes such as STL7, TGFβR3, TNFRSF25 and CHD5. Deletions on 22q and 3q occurred in approximately 50% of pheochromocytoma samples and three respectively two MORs were described on these autosomes. Additional common deletions (≥=32%) occurred on chromosome arms 11p, 11q and 17p while 1q, 7p, 12q, and 19p harboured the most common MORs of gain (≥=14%). Our results confirmed previous reports which predominantly used lower resolution methods, and importantly, revealed several recurrent and narrow, as well as novel, chromosomal regions likely to be associated to the development of pheochromocytoma and paraganglioma. The finding of more gains in the malignant pheochromocytomas compared to the benign cases is noteworthy and suggestive of the contributions of overexpression of oncogenes in the progression to malignancy.

In conclusion, the identification of candidate regions in both benign and malignant tumours facilitates the discovery of tumour suppressor genes and oncogenes with implications for the pathogenesis of adrenal and extra- adrenal catecholamine producing tumours.

60 Paper IV

Integrative Epigenomic and Genomic Analysis of Malignant Pheochromocytoma Histone modifications are epigenomic regulators known to affect gene expression and H3K4me3 and H3K27me3 are two relatively well-studied histone marks associated with active and silent genes, respectively. Epigenomic mechanisms have not however been investigated extensively in pheochromocytomas so far. Genomic aberrations, such as DNA copy number changes are common in tumours and also affect gene transcription levels.

Our epigenomic survey of the H3K4me3 and H3K27me3 in a malignant pheochromocytoma led to the detection of 16,096 and 27,694 enriched regions respectively. We also identified 2,806 sites of overlapping H3K4me3 and H3K27me3 enrichment, i.e. bivalent regions within the sample. In concordance with previous studies, H3K4me3 was found as a frequent mark in the promoter regions of both known and presumably novel transcripts while H3K27me3 seemed mainly to be located at intergenic and/or intragenic locations further down in the gene body. Gene Ontology analysis revealed that H3K4me3-associated genes were significantly over-represented in house- keeping and metabolic processes. H3K27me3- and bivalently-associated genes were, on the other hand, significantly over-represented with developmental process terms. When integrating the determined gene expression and DNA copy number data (Paper III), we first observed an expected significant relationship between H3K4me3-associated genes and relatively high expression as well as a relation of H3K27me3-associated genes and low-expressed genes. Secondly, the corresponding significant trend was found for the groups of high- and low-expressed genes and their mapping to genomic regions of gains and losses. A comparison of the expression levels in the tumour sample with those in normal adrenal medulla revealed approximately 2,500 differentially expressed genes. For genes with increased expression an over-representation of genes located in gained regions as well as an under- representation of genes in deleted regions were found. The opposite relations were found for genes with decreased expression. However, for the histone modification data, only genes with decreased expression showed significant over- and under-representation of H3K27me and H3K4me3 enrichment, respectively. This implied a transition of an active histone mark to a silent one for a number of genes in the present sample. Genes with bivalent marks showed in general decreased expression in the tumour. Thus, the analyses indicated that either histone modifications or chromosomal alterations, or both, may be the underlying responsible cause of the altered expression for a

61 substantial fraction of genes in the investigated malignant pheochromocytoma.

Candidate tumour suppressor genes with decreased expression, a H3K27me3 mark and/or located in regions of deletion were for instance TGIF1, DSC3, TNFRSF10B, RASSF2, HOXA9, PTPRE and CDH11. More genes were found with increased expression, a H3K4me3 mark and/or in regions of gain. Potential oncogenes detected among those were GNAS, INSM1, DOK5, ETV1, RET, NTRK1 and IGF2. Of great interest was the finding that the polycomb associated trimethylase gene EZH2 showed increased expression. EZH2 was also situated in a region with increase in DNA copy number and was semi-bivalently marked (non-overlapping H3K4me3 and H3k27me enrichment). The up-regulation of EZH2 mRNA production has been observed in several other cancers suggestive of an important role for H3K27me3 in tumourigenesis and as a potential biomarker. This result may also relate to the high number of enriched regions of H3K27me3 found in the investigated sample and the over-representation of H3K27me3 marks at the down-regulated genes. However, no histone modification map for normal adrenal medulla is available.

Gene interaction network analysis was performed to investigate the biological relationships among genes with aberrant expression. This revealed that the most enriched network for genes with increased expression in the tumour was related to cell morphology and cell transport. Also, when adding both histone modifications and DNA copy number data for the interacting genes it was apparent that these changes might contribute to the increased expression. This network included the genes RET, DOK5 and INSM1 among others. Interestingly, a gene network related to apoptosis was identified for the genes displaying decreased expression in the pheochromocytoma. This gene network indicated interactions between TNFRSF10B, FAD, CASP4, CASP8 and several other genes involved in apoptosis.

To conclude, the integrated analysis of the epigenomic and genomic statuses in this malignant pheochromocytoma has shown the association of histone modifications and DNA copy number changes to differentially expressed genes and revealed their apparent impact on global gene transcription. Moreover, the approach enabled the identification of interesting individual candidate tumour genes.

62 CONCLUDING REMARKS

Ongoing research and future perspectives

In this thesis, I have described the results of extensive studies both at the genomic and epigenomic level of normal and tumour cells. The initial and main focus of the present investigations has been on DNA copy number changes, both in the context of apparently healthy cells, as well as the identification of chromosomal aberrations linked to the development of tumours in endocrine organs.

CNVs have in recent years been recognized as a predominant source of human genetic variation. However, further investigations revealing the distribution and frequency of this type of structural variant are needed both in health and disease to increase our knowledge about the pathogenic associated CNVs. In paper I we established and applied a 32K array which was also utilized in subsequent studies. The work in this initial paper created a valuable resource of CNV information in healthy individuals. Future research aiming to discover CNVs along with the elucidation of their impact on phenotype and evolution, among other questions that remain to be answered, will probably be done by applying the massive parallel sequence technology and large sample sets, which enables extreme high resolution studies as well as detection of rare variants. However within the clinic the less expensive microarray platforms may be increasingly used in the years to come, aiding in the diagnosis of many different human traits which are due to DNA copy number changes. Further research will reveal the full impact of CNVs on gene expression and the identification of variants with strong associations to disease and possibly also point to genes suitable for targeted therapy.

CNV in the context of somatic mosaicism in healthy differentied tissues, addressed in paper II, is an understudied topic and future research will hopefully reveal more about its´ extent, rate of formation and the importance of genetically distinct populations of cells in the same organisms. The use of more sensitive methodological approaches will also assist the detection of a small fraction of cells carrying different genomic profiles. The importance of somatic mosiacism may be greater than presently acknowledged since when

63 studying the development of sporadic disorders, the comparison to the cor- rect matching control tissue will be very relevant.

Paper III and IV concerns the genomics and epigenomics of pheochromocytomas, where the DNA copy number profiling of 53 tumours in paper III represents to date the most comprehensive high resolution study of benign and malignant pheochromocytomas and paragangliomas. This genome wide survey identified several interesting chromosomal regions of losses and gains where several of them probable are significant both for the development and the progression of these catecholamine producing tumours. We are currently following up a few candidate genes located in some of these regions. Among these, we selected HIC1 (Hypermethylated in Cancer 1) which is a candidate tumour suppressor gene located at 17p13.3. Approximately 35% of the studied tumours in paper III presented with deletion of the telomeric part of chromosome arm 17p, implying impact on the tumourgenesis. The HIC1 locus is often subjected to chromosomal loss also in several other human tumour forms. Moreover, as the name suggests, this genes is frequently found hypermethylated in cancers (277). HIC1 encodes a sequence-specific transcriptional repressor and is positively regulated by p53 and other p53 members (278). The gene seems to contain three promoters that give rise to alternatively spliced transcripts. In our present research we have investigated the CpG methylation status at the promotor for one of the common transcripts which has been found to be densely methylated in many cancer cells. This methylation pattern has also been shown to correlate with lack of expression of the HIC1 gene (278). An earlier publication using methylation- specific PCR reported methylation of this gene in a few sporadic pheochromocytomas (263). We have extended the analysis by investigating the HIC1 methylation status in, so far 22 benign and malignant pheochromocytomas with the quantitative method of pyrosequencing. Preliminary results from the average values of 6 consecutive CpGs show variable level of methylation, ranging from 9-58% in the tumours. We will furthermore consider the DNA copy number data, as well as measure the HIC1 expression. Altogether this will allow us to substantiate a possible tumour suppressor role of HIC1 in pheochromocytoma.

Loss at 1p36 was a highly frequent event in the studied series of tumours in paper III and furthermore, deletions at this chromosomal band on this autosome have been reported earlier in pheochromocytomas as well as in various other cancers, such as neuroblastoma (279) . These findings suggest that inactivation of a tumour suppressor gene(s) mapping to this region may be an initial event. Many candidate genes locate at 1p36 and one of the more interesting ones is the CHD5 gene. CHD5 is also included in one of the four MORs of deletions identified in paper III. Of the 53 investigated pheochro-

64 mocytomas and paragangliomas, 33 (62%) presented with loss at the CHD5 locus. Chromodomain helicase DNA-binding protein 5 (CHD5) has been proposed to function as a chromatin remodeling protein. Functional studies including knock down experiments have recently also shown the tumour suppressive effects of CHD5 (279). Hypermethylation of the CHD5 promotor region has been reported in several human primary tumours and cancer cell lines, for instance neuroblastoma, which also correlated with low or absent expression (280). We have therefore performed a preliminary screening of the methylation status in a few pheochomcytomas by bisulphite sequencing a region containing 35 CpGs close to the promoter of the CHD5 gene (the same regions as reported in 281). Our results however showed a very low amount of DNA methylation. Further studies to elucidate the importance of the CHD5 candidate tumour suppressor gene should include measurement of expression levels and mutation analysis.

Several additional interesting genes are located in the defined MORs of frequent deletions on chromosome arm 1p and one approach could be to sequence the exons of the genes encompassed within those. The MOR with smallest size at 1p36, a 3.29Mb region that is lost in 63% of tumours, is an attractive candidate. This region also harbors the CHD5 gene among 25 others. Even more frequently deleted, in 76% of analysed pheochromocytomas, is the 4.73Mb sized MOR at 1p13 with a total of 60 genes, of which many present with potential tumour suppressor functions. The exonic regions representing the genes in these MORs in several pheochromocytoma samples could be subjected for selected enrichment/capturing with available array-based methods or in-solution alternatives followed by sequencing using next-generation techniques (massive parallel), for detection of possible disease causing mutations.

Recent technical developments have also made it feasible to enrich for all the protein-coding regions in the human genome in one experiment. These regions constitute ~1% of the genome or ~30Mb split across ~180 000 exons. The fact that approximately 85% of disease causing mutations are typically found in the coding region or splice sites (282) makes this a promising approach to reduce costs and discover mutations of importance in the studied samples until whole genome sequencing becomes more routinely performed. We will continue to explore the genetics of apparently sporadic pheochromcytomas and utilize an array-based capture assay (Nimblegene) for this purpose on initially a limited numbers of tumours, followed by massive parallel sequencing (Illumina) to search the entire exome for somatic mutations. Constitutional DNA from the same patients will also be sequenced to distinguish predisposing mutations from somatic mutations.

65 The integrative epigenomic study of the malignant pheochromocytoma sample in paper IV uncovered the global map of the histone modifications H3K4me3 and H3K27me3 as well as their association to gene expression. With the addition of DNA copy number data, both general and gene-specific mechanistic relations were found for high- and low-expressed genes as well as for differentially expressed genes to either or both the histone marks and the copy number statuses. Several tumour biologically interesting gene candidates were found up- or down-regulated with an associated histone modification and/or copy number change. For instance the PRC2 component EZH2 which functions as a H3K27 methyltransferase was found up regulated in the present sample. This potential oncogene has been reported to be over expressed in many cancers. The gene silencing effect of H3K27 trimethylation, mediated by EZH2 and the association of this histone mark to known tumour suppressor genes imply its importance in cancer (166) and also suggest for future novel findings of genes aberrantly silenced in tumours via this mechanism. We have initiated studies to determine the overall relative expression status of EZH2 in benign and malignant pheochromocytoma. Future experiments also include treatment of primary pheochromocytoma cells with 3-Deazeneplanocin (DZNep), a histone methylation inhibitor that is known to efficiently deplete EZH2 in other cell types. Treatment of pheochromocytoma cells with DZNep may result in reactivation of H3K27me3-repressed target genes, reduced cell viability, and induction of apoptosis as have been reported to occur for other tumour cells (283).

General and future genomic and epigenomic views on pheochromocytoma and cancer There are particular aspects that are highly relevant in the context of epigenetic mechanisms and gene regulation in cancer research. A better understanding is obviously desirable but also the fact that these epigenetic marks are in principle reversible, as opposed to genetic changes. It should be possible, at least theoretically to counteract the tumour-specific epigenetic alterations, leading the cancer cell to death or differentation. Epigenetic mechanisms include specific enzymes whose activity could be antagonized by small molecule inhibitors and are therefore represent candidate drug targets. Two well known pharmacological agents for the reversal of DNA methylation are also approved for therapeutic use in leukemia, 5-aza-cytidine and 5-aza-2’- deoxycytidine. Most compounds affecting the epigenetic status of cells, including histone acetylation and methylation, are still in experimental phase, such as DZNep which has shown cancer cell-specific induced apoptosis (284, 285). With increased specificity and positive evaluations of developed

66 agents, these therapeutic strategies against the epigenome of cancer cells will hopefully reach the clinic in the near future.

Progress is however dependent on more extensive epigenomic profiling and analysis to reveal the tumour specific aberrations. Such research is thus increasingly produced with the help of technical advances to globally investigate for epigenetic changes as well as more locally in larger samples set, which gives promises for the identification of both prognostic patterns and suitable drug candidates. Explorations of this kind will be of value to perform on pheochromocytomas, since the epigenome remains to be established.

Improved knowledge at the genetic and genomic level is also greatly needed for these neuroendocrine tumours and the combination of higher resolution surveys of DNA copy number changes , global gene expression profiling as well as whole genome/exome sequencing with epigenomic analysis will produce exciting results. Such large scale integrative studies addressing both individual genes and affected signaling pathways, may very likely aid in the identification of additional genes important in the development of the sporadic cases with today unknown cause. Moreover, genetic and epigenetic examinations of large collections of malignant pheochromocytomas in comparison to benign cases are also attractive to perform due to the present lack of reliable markers for prediction of malignancy. Notably, regulatory miR- NAs may be of importance for progression as was suggested in a recent study (286). Additionally increased knowledge of the pathways linked to pheochromocytoma such as the HIF-1 related signaling transduction pathway, which regulates genes involved in angiogenesis, cell survival and metastasis would be valuable. In recent years many therapies have been developed to target angiogenesis in cancer and since many malignant pheochromocytomas and paragangliomas are associated with excessive angiogenesis patients may benefit from drugs that target VEGF and/or PDGF-beta receptors. These proteins promote angiogenesis and are under the control of HIF-1 and are subsequently often upregulated in these tumours. There are also a few case reports on patients treated with the potent tyrosine kinase inhibitor Suntinib, which targets VEGFR and PDGFR among other tyrosine kinase receptors, with promising results of tumour shrinkage (195). Other approaches discussed in the literature include molecules that inhibit HIF-1 activity directly or indirectly, however, more is needed to be learned about the complexity of the HIF-1 pathways and the consequences of modulating that. ERBB-2, another tyrosine kinase receptor also mentioned earlier as upregulated in malignant pheochromocytomas is a also potential target for therapy. ERBB-2 signaling also often increases HIF-1 and subsequently VEGF expression, making the study of the clinical effect of the existing

67 ERBB-2 inhibitors desirable to perform (195). Further research and careful evaluation of promising candidates for future targeted therapies are required.

Finally, applicable to tumours in general, even though cancer genomes seem to be very heterogeneous, yet in the same subtypes of tumours the affected signaling pathways might be the same across the majority of samples as discussed above, and drugs are therefore being developed to target those pathways (287). Optimistic is also the advances of new and less costly DNA sequencing techniques enabling for individual screening of cancer patients for the best possible diagnosis. Even more cost efficient could be the use of validated tumour-specific diagnostic and/or prognostic arrays based on known and imminent genetic data. With the rapid increase of knowledge in the field of cancer research and availability of varous specific drugs against driver genes, personalized treatment will hopefully soon be a reality for the majority of cancer patients.

68 ACKNOWLEDGMENTS

This work was jointly carried out at the Department of Genetics and Pathol- ogy, Uppsala University and the Department of Surgical Sciences, Uppsala University Hospital. Financial funding was provided by the Swedish Cancer Society, the Swedish Research Council, Swedish Children Cancer Founda- tion, Letterstedtska föreningen and Lion´s Cancer Research Foundation, Uppsala.

I would like to express my sincere gratitude to a number of important people that in various ways have been a support during the years as a PhD student, which resulted in this thesis, among other things.

Professor Gunnar Westin, my main supervisor. I am grateful for your con- stant encouragement, sharing of vast knowledge and enthusiasm for science. I would also like to thank you for giving me the opportunity to work relatively independent at the Rudbeck lab. However, with regular, and whenever needed, constructive discussions about my research. It has been a privilege being your PhD student.

Associate professor Teresita Díaz de Ståhl my co-supervisor, who has been both greatly inspiring and helpful to me during these years. I am extremely impressed by your capability in research and I am also thankful for the con- fidence in this area that you have given me. I look forward to continue colla- borate with you in the future!

My co-supervisor professor Göran Åkerström for accepting me as a PhD student at the Department of Surgical Sciences and for always being suppor- tive and able to find time for valuable contribution regarding your immense expertise in the field of endocrine tumours. I am grateful for this period in the proximity of the excellent clinical and research acitivities conducted under your supervison.

Ola Hessman, PhD, my third co-supervisor. Thank you for all assistance, mainly concerning clinical data.

Professor Jan Dumanski, for letting me do a major part of this work in your lab and introducing me to the field of genomic microarrays as well as shar-

69 ing great knowledge. I have really enjoyed it as well as learnt a lot due to all discussions about science, all the reading of manuscripts and applications and not at least due to the ambitious environment in the group. Carl Bruder, PhD, I missed you the very last years. I really appreciated working with you during my initial PhD –period, among many things; for your positive attitude and the excellent supervision you gave me. Thank you also for the immense hospitality during my visits in Alabama at University of Alabama at Bir- mingham (UAB). I look forward to keep in contact (and who knows, maybe not only science needs to be discussed then but also some practical chopper battle can be conducted?!). Huge thanks also to Desiree von T for opening your beautiful home to me in Birmingham!

My co-authors and collaborators in Uppsala, Stockholm, USA, Germany, Poland and elsewhere, in particular Robin Andersson at LCB for program- ming and bioinformatic skills of great significance for all our projects. Thank you for productive collaboration and for always trying to realize all my suggestions of additional data analysis. Professor Claes Wadelius for initiating the interesting and extensice collaboratoin concerning cancer epigenomics . I would also like to thank his former PhD students, Alvaro Ra- da-Iglesias for invaluable experimental and theoretical help as well as Meh- di Motallebipour for his assistance . Professor Catharina Larsson, at Ka- rolinska Institutet, regarding the sucessful pheochromocytoma/paraganglioma project. Professor Per-Uno Malmström for bladder cancer collaboration.

Present and past members of the Westin/Åkerström group, especially Jessica Svedlund, queen of the wetlab! and with impressive multitasking skills (due to all the students?), thank you for the kind guidance at your lab. Alberto Delgado and Katarina Edfeldt, for your fun company in Frankfurt among other locations and things, and good luck both with the array-studies! Birgit- ta Bondesson, thanks for all the help with all the tumour sections. Peter Lillhager, for your expertise in various molecular biology methods. Peyman Björklund, PhD, for your inspiring research interest, ideas and make-it- happen. Ulrika Segersten, PhD,for continuous collaborations at Rudbeck as well. Professor Per Hellman and associate professor Peter Stålberg for valuable discussions and collaborations.

Past and present members of Dumanski/Diaz de Ståhl groups, especially Helena Nord, PhD, it has been great to have you as roommate, lunch-mate, “vi-röstar-rött”-mate and so forth. Good luck with everything and let´s keep in touch! Arek Piotrowski, PhD, thanks for your sharing of scientific knowledge, both in Sweden, in USA and from Poland. Cihan Cetinkaya, PhD, for also being such a friendly person! Uwe, Caisa, Cecilia, Patrick, Magda, Caroline, Ricardo, Kiran, Birgitta (du flyttar väl inte till Argentina snart,

70 eller?) and other previous members that I have enjoyed working with. Thanks also to the present people in the lab, Hamid Reza Razzaghian, good luck with your PhD! Lars Forsberg, PhD, I appreciate all our fun talks (“öronvax, kan man använda det för arrayanalys”), Geeta and Chiara.

All other people working at Rudbeck and at Klinisk Forskningsavdelning 2 for making it pleasant, particulary “innebandy-gänget” from IGP and elsewhere, I will miss acting out my frustrations on Fridays. Also: Lotta T and Magnus S for involvement in array and pyrosequenincing projects respectively, and all the members in the Chromatin journal club! as well as Janet C, for fruitful discussions. Obviously also a huge number of additional great persons and researcher at these locations but I have to stop counting some- where..

I would moreover like to express my gratitude to all the patients and controls that have participated in the performed studies.

Alla tidigare kursare och lärare vid Uppsala Universitet, speciellt Helena Å och Niklas N, snart är vi doktorer alla 3!

Jag vill även här tacka min familj och släkt, samt mina vänner för att ni fyl- ler livet mitt med så mycket viktigt utanför forskningsarbetet. Som konserter och rediga musikfestivaler tex, eller hur Stina och Marcus? Jag uppskattar er vänskap och alla trevligheter vi ägnar oss åt ihop, hör ni det allihopa, ni vet vilka ni är! I synnerhet min vapendragare Moa, du är bäst. Precis som Lisa, Annika, Ida och ni andra sköna och fina människor jag känner, i bla Uppsala, Stockholm, Gävle och Berlin! Inklusive spelare i det eminenta kor- penlaget BK Fimpen.

Mina föräldrar, Jan och Margareta, för oskattbar trygghet och det stöd jag alltid får från er, gällande alla möjliga hänseenden. Ni båda är nog mer bety- delsefulla för mig än vad ni förstår, tack för allt. Inbegripet era tidvis tappra försök att engagera er i min forskning, bara att börja läsa nu, ni har en relati- vit ny och välskriven sammanfattning just här. Jag lovar dock som vanligt att uppdatera er om kommande projekt. Min bror Erik, den bästa lillebror man kan ha. Våra likheter och olikheter gör det hela ännu bättre, jag ser fram emot att komma och hälsa på hos dig och Helena i Malmö, igen, och igen...

Resterande släkt, särskilt Johnny och Maud med barn och deras familjer, erat genuina intresse för min avhandling har betytt mycket. Samt ”svär- föräldrarna” Olle och Sissi på vackra Risö.

Min Lasse/o, du briljanta själsfrände (+/- lite svensk litteraturhistoria, base- ball och naturvetenskap men annars så!), allt som är du är kärlek.

72 REFERENCES

1. Watson JD, Crick FH. Molecular structure of nucleic acids; a structure for deoxyribose nucleic acid. Nature 1953;171(4356):737-8. 2. Amaral PP, Dinger ME, Mercer TR, Mattick JS. The eukaryotic genome as an RNA machine. Science 2008;319(5871):1787-9. 3. Lander ES, Linton LM, Birren B, Nusbaum C, Zody MC, Baldwin J, Devon K, Dewar K, Doyle M, FitzHugh W, Funke R, Gage D, Harris K, Heaford A, Howland J, Kann L, Lehoczky J, LeVine R, McEwan P, McKernan K, Meldrim J, Mesirov JP, Miranda C, Morris W, Nay- lor J, Raymond C, Rosetti M, Santos R, Sheridan A, Sougnez C, Stange-Thomann N, Stojanovic N, Subramanian A, Wyman D, Rogers J, Sulston J, Ainscough R, Beck S, Bentley D, Burton J, Clee C, Cart- er N, Coulson A, Deadman R, Deloukas P, Dunham A, Dunham I, Durbin R, French L, Grafham D, Gregory S, Hubbard T, Humphray S, Hunt A, Jones M, Lloyd C, McMurray A, Matthews L, Mercer S, Milne S, Mullikin JC, Mungall A, Plumb R, Ross M, Shownkeen R, Sims S, Waterston RH, Wilson RK, Hillier LW, McPherson JD, Mar- ra MA, Mardis ER, Fulton LA, Chinwalla AT, Pepin KH, Gish WR, Chissoe SL, Wendl MC, Delehaunty KD, Miner TL, Delehaunty A, Kramer JB, Cook LL, Fulton RS, Johnson DL, Minx PJ, Clifton SW, Hawkins T, Branscomb E, Predki P, Richardson P, Wenning S, Slezak T, Doggett N, Cheng JF, Olsen A, Lucas S, Elkin C, Uberbacher E, Frazier M, Gibbs RA, Muzny DM, Scherer SE, Bouck JB, Sodergren EJ, Worley KC, Rives CM, Gorrell JH, Metzker ML, Naylor SL, Ku- cherlapati RS, Nelson DL, Weinstock GM, Sakaki Y, Fujiyama A, Hattori M, Yada T, Toyoda A, Itoh T, Kawagoe C, Watanabe H, To- toki Y, Taylor T, Weissenbach J, Heilig R, Saurin W, Artiguenave F, Brottier P, Bruls T, Pelletier E, Robert C, Wincker P, Smith DR, Dou- cette-Stamm L, Rubenfield M, Weinstock K, Lee HM, Dubois J, Ro- senthal A, Platzer M, Nyakatura G, Taudien S, Rump A, Yang H, Yu J, Wang J, Huang G, Gu J, Hood L, Rowen L, Madan A, Qin S, Davis RW, Federspiel NA, Abola AP, Proctor MJ, Myers RM, Schmutz J, Dickson M, Grimwood J, Cox DR, Olson MV, Kaul R, Raymond C, Shimizu N, Kawasaki K, Minoshima S, Evans GA, Athanasiou M, Schultz R, Roe BA, Chen F, Pan H, Ramser J, Lehrach H, Reinhardt

73 R, McCombie WR, de la Bastide M, Dedhia N, Blocker H, Hornischer K, Nordsiek G, Agarwala R, Aravind L, Bailey JA, Bateman A, Bat- zoglou S, Birney E, Bork P, Brown DG, Burge CB, Cerutti L, Chen HC, Church D, Clamp M, Copley RR, Doerks T, Eddy SR, Eichler EE, Furey TS, Galagan J, Gilbert JG, Harmon C, Hayashizaki Y, Haussler D, Hermjakob H, Hokamp K, Jang W, Johnson LS, Jones TA, Kasif S, Kaspryzk A, Kennedy S, Kent WJ, Kitts P, Koonin EV, Korf I, Kulp D, Lancet D, Lowe TM, McLysaght A, Mikkelsen T, Moran JV, Mulder N, Pollara VJ, Ponting CP, Schuler G, Schultz J, Slater G, Smit AF, Stupka E, Szustakowski J, Thierry-Mieg D, Thier- ry-Mieg J, Wagner L, Wallis J, Wheeler R, Williams A, Wolf YI, Wolfe KH, Yang SP, Yeh RF, Collins F, Guyer MS, Peterson J, Fel- senfeld A, Wetterstrand KA, Patrinos A, Morgan MJ, de Jong P, Ca- tanese JJ, Osoegawa K, Shizuya H, Choi S, Chen YJ. Initial sequencing and analysis of the human genome. Nature 2001;409(6822):860- 921. 4. Venter JC, Adams MD, Myers EW, Li PW, Mural RJ, Sutton GG, Smith HO, Yandell M, Evans CA, Holt RA, Gocayne JD, Amanatides P, Ballew RM, Huson DH, Wortman JR, Zhang Q, Kodira CD, Zheng XH, Chen L, Skupski M, Subramanian G, Thomas PD, Zhang J, Ga- bor Miklos GL, Nelson C, Broder S, Clark AG, Nadeau J, McKusick VA, Zinder N, Levine AJ, Roberts RJ, Simon M, Slayman C, Hunka- piller M, Bolanos R, Delcher A, Dew I, Fasulo D, Flanigan M, Florea L, Halpern A, Hannenhalli S, Kravitz S, Levy S, Mobarry C, Reinert K, Remington K, Abu-Threideh J, Beasley E, Biddick K, Bonazzi V, Brandon R, Cargill M, Chandramouliswaran I, Charlab R, Chaturvedi K, Deng Z, Di Francesco V, Dunn P, Eilbeck K, Evangelista C, Ga- brielian AE, Gan W, Ge W, Gong F, Gu Z, Guan P, Heiman TJ, Hig- gins ME, Ji RR, Ke Z, Ketchum KA, Lai Z, Lei Y, Li Z, Li J, Liang Y, Lin X, Lu F, Merkulov GV, Milshina N, Moore HM, Naik AK, Na- rayan VA, Neelam B, Nusskern D, Rusch DB, Salzberg S, Shao W, Shue B, Sun J, Wang Z, Wang A, Wang X, Wang J, Wei M, Wides R, Xiao C, Yan C, Yao A, Ye J, Zhan M, Zhang W, Zhang H, Zhao Q, Zheng L, Zhong F, Zhong W, Zhu S, Zhao S, Gilbert D, Baumhueter S, Spier G, Carter C, Cravchik A, Woodage T, Ali F, An H, Awe A, Baldwin D, Baden H, Barnstead M, Barrow I, Beeson K, Busam D, Carver A, Center A, Cheng ML, Curry L, Danaher S, Davenport L, Desilets R, Dietz S, Dodson K, Doup L, Ferriera S, Garg N, Glueck- smann A, Hart B, Haynes J, Haynes C, Heiner C, Hladun S, Hostin D, Houck J, Howland T, Ibegwam C, Johnson J, Kalush F, Kline L, Ko- duru S, Love A, Mann F, May D, McCawley S, McIntosh T, McMul- len I, Moy M, Moy L, Murphy B, Nelson K, Pfannkoch C, Pratts E, Puri V, Qureshi H, Reardon M, Rodriguez R, Rogers YH, Romblad D, Ruhfel B, Scott R, Sitter C, Smallwood M, Stewart E, Strong R, Suh E, Thomas R, Tint NN, Tse S, Vech C, Wang G, Wetter J, Williams S,

74 Williams M, Windsor S, Winn-Deen E, Wolfe K, Zaveri J, Zaveri K, Abril JF, Guigo R, Campbell MJ, Sjolander KV, Karlak B, Kejariwal A, Mi H, Lazareva B, Hatton T, Narechania A, Diemer K, Muruganu- jan A, Guo N, Sato S, Bafna V, Istrail S, Lippert R, Schwartz R, Wa- lenz B, Yooseph S, Allen D, Basu A, Baxendale J, Blick L, Caminha M, Carnes-Stine J, Caulk P, Chiang YH, Coyne M, Dahlke C, Mays A, Dombroski M, Donnelly M, Ely D, Esparham S, Fosler C, Gire H, Glanowski S, Glasser K, Glodek A, Gorokhov M, Graham K, Grop- man B, Harris M, Heil J, Henderson S, Hoover J, Jennings D, Jordan C, Jordan J, Kasha J, Kagan L, Kraft C, Levitsky A, Lewis M, Liu X, Lopez J, Ma D, Majoros W, McDaniel J, Murphy S, Newman M, Nguyen T, Nguyen N, Nodell M, Pan S, Peck J, Peterson M, Rowe W, Sanders R, Scott J, Simpson M, Smith T, Sprague A, Stockwell T, Turner R, Venter E, Wang M, Wen M, Wu D, Wu M, Xia A, Zandieh A, Zhu X. The sequence of the human genome. Science 2001;291(5507):1304-51. 5. Ku CS, Loy EY, Salim A, Pawitan Y, Chia KS. The discovery of human genetic variations and their use as disease markers: past, present and future. Journal of human genetics 2010;55(7):403-15. 6. HapMap. The International HapMap Project. Nature 2003;426(6968):789-96. 7. Levy S, Sutton G, Ng PC, Feuk L, Halpern AL, Walenz BP, Axelrod N, Huang J, Kirkness EF, Denisov G, Lin Y, MacDonald JR, Pang AW, Shago M, Stockwell TB, Tsiamouri A, Bafna V, Bansal V, Kra- vitz SA, Busam DA, Beeson KY, McIntosh TC, Remington KA, Abril JF, Gill J, Borman J, Rogers YH, Frazier ME, Scherer SW, Strausberg RL, Venter JC. The diploid genome sequence of an individual human. PLoS Biol 2007;5(10):e254. 8. Bentley DR, Balasubramanian S, Swerdlow HP, Smith GP, Milton J, Brown CG, Hall KP, Evers DJ, Barnes CL, Bignell HR, Boutell JM, Bryant J, Carter RJ, Keira Cheetham R, Cox AJ, Ellis DJ, Flatbush MR, Gormley NA, Humphray SJ, Irving LJ, Karbelashvili MS, Kirk SM, Li H, Liu X, Maisinger KS, Murray LJ, Obradovic B, Ost T, Par- kinson ML, Pratt MR, Rasolonjatovo IM, Reed MT, Rigatti R, Rodig- hiero C, Ross MT, Sabot A, Sankar SV, Scally A, Schroth GP, Smith ME, Smith VP, Spiridou A, Torrance PE, Tzonev SS, Vermaas EH, Walter K, Wu X, Zhang L, Alam MD, Anastasi C, Aniebo IC, Bailey DM, Bancarz IR, Banerjee S, Barbour SG, Baybayan PA, Benoit VA, Benson KF, Bevis C, Black PJ, Boodhun A, Brennan JS, Bridgham JA, Brown RC, Brown AA, Buermann DH, Bundu AA, Burrows JC, Carter NP, Castillo N, Chiara ECM, Chang S, Neil Cooley R, Crake NR, Dada OO, Diakoumakos KD, Dominguez-Fernandez B, Earn- shaw DJ, Egbujor UC, Elmore DW, Etchin SS, Ewan MR, Fedurco M, Fraser LJ, Fuentes Fajardo KV, Scott Furey W, George D, Gietzen KJ, Goddard CP, Golda GS, Granieri PA, Green DE, Gustafson DL,

75 Hansen NF, Harnish K, Haudenschild CD, Heyer NI, Hims MM, Ho JT, Horgan AM, Hoschler K, Hurwitz S, Ivanov DV, Johnson MQ, James T, Huw Jones TA, Kang GD, Kerelska TH, Kersey AD, Khreb- tukova I, Kindwall AP, Kingsbury Z, Kokko-Gonzales PI, Kumar A, Laurent MA, Lawley CT, Lee SE, Lee X, Liao AK, Loch JA, Lok M, Luo S, Mammen RM, Martin JW, McCauley PG, McNitt P, Mehta P, Moon KW, Mullens JW, Newington T, Ning Z, Ling Ng B, Novo SM, O'Neill MJ, Osborne MA, Osnowski A, Ostadan O, Paraschos LL, Pickering L, Pike AC, Pike AC, Chris Pinkard D, Pliskin DP, Pod- hasky J, Quijano VJ, Raczy C, Rae VH, Rawlings SR, Chiva Rodri- guez A, Roe PM, Rogers J, Rogert Bacigalupo MC, Romanov N, Ro- mieu A, Roth RK, Rourke NJ, Ruediger ST, Rusman E, Sanches- Kuiper RM, Schenker MR, Seoane JM, Shaw RJ, Shiver MK, Short SW, Sizto NL, Sluis JP, Smith MA, Ernest Sohna Sohna J, Spence EJ, Stevens K, Sutton N, Szajkowski L, Tregidgo CL, Turcatti G, Vande- vondele S, Verhovsky Y, Virk SM, Wakelin S, Walcott GC, Wang J, Worsley GJ, Yan J, Yau L, Zuerlein M, Rogers J, Mullikin JC, Hurles ME, McCooke NJ, West JS, Oaks FL, Lundberg PL, Klenerman D, Durbin R, Smith AJ. Accurate whole human genome sequencing using reversible terminator chemistry. Nature 2008;456(7218):53-9. 9. Wheeler DA, Srinivasan M, Egholm M, Shen Y, Chen L, McGuire A, He W, Chen YJ, Makhijani V, Roth GT, Gomes X, Tartaro K, Niazi F, Turcotte CL, Irzyk GP, Lupski JR, Chinault C, Song XZ, Liu Y, Yuan Y, Nazareth L, Qin X, Muzny DM, Margulies M, Weinstock GM, Gibbs RA, Rothberg JM. The complete genome of an individual by massively parallel DNA sequencing. Nature 2008;452(7189):872- 6. 10. Wang J, Wang W, Li R, Li Y, Tian G, Goodman L, Fan W, Zhang J, Li J, Zhang J, Guo Y, Feng B, Li H, Lu Y, Fang X, Liang H, Du Z, Li D, Zhao Y, Hu Y, Yang Z, Zheng H, Hellmann I, Inouye M, Pool J, Yi X, Zhao J, Duan J, Zhou Y, Qin J, Ma L, Li G, Yang Z, Zhang G, Yang B, Yu C, Liang F, Li W, Li S, Li D, Ni P, Ruan J, Li Q, Zhu H, Liu D, Lu Z, Li N, Guo G, Zhang J, Ye J, Fang L, Hao Q, Chen Q, Liang Y, Su Y, San A, Ping C, Yang S, Chen F, Li L, Zhou K, Zheng H, Ren Y, Yang L, Gao Y, Yang G, Li Z, Feng X, Kristiansen K, Wong GK, Nielsen R, Durbin R, Bolund L, Zhang X, Li S, Yang H, Wang J. The diploid genome sequence of an Asian individual. Nature 2008;456(7218):60-5. 11. Frazer KA, Murray SS, Schork NJ, Topol EJ. Human genetic variation and its contribution to complex traits. Nat Rev Genet 2009;10(4):241- 51. 12. Schlotterer C. The evolution of molecular markers--just a matter of fashion? Nat Rev Genet 2004;5(1):63-9. 13. Nakamura Y. DNA variations in human and medical genetics: 25 years of my experience. Journal of human genetics 2009;54(1):1-8.

76 14. Iafrate AJ, Feuk L, Rivera MN, Listewnik ML, Donahoe PK, Qi Y, Scherer SW, Lee C. Detection of large-scale variation in the human genome. Nat Genet 2004;36(9):949-51. 15. Sebat J, Lakshmi B, Troge J, Alexander J, Young J, Lundin P, Maner S, Massa H, Walker M, Chi M, Navin N, Lucito R, Healy J, Hicks J, Ye K, Reiner A, Gilliam TC, Trask B, Patterson N, Zetterberg A, Wigler M. Large-scale copy number polymorphism in the human genome. Science 2004;305(5683):525-8. 16. Tuzun E, Sharp AJ, Bailey JA, Kaul R, Morrison VA, Pertz LM, Hau- gen E, Hayden H, Albertson D, Pinkel D, Olson MV, Eichler EE. Fine-scale structural variation of the human genome. Nat Genet 2005;37(7):727-32. 17. Redon R, Ishikawa S, Fitch KR, Feuk L, Perry GH, Andrews TD, Fiegler H, Shapero MH, Carson AR, Chen W, Cho EK, Dallaire S, Freeman JL, Gonzalez JR, Gratacos M, Huang J, Kalaitzopoulos D, Komura D, MacDonald JR, Marshall CR, Mei R, Montgomery L, Ni- shimura K, Okamura K, Shen F, Somerville MJ, Tchinda J, Valsesia A, Woodwark C, Yang F, Zhang J, Zerjal T, Zhang J, Armengol L, Conrad DF, Estivill X, Tyler-Smith C, Carter NP, Aburatani H, Lee C, Jones KW, Scherer SW, Hurles ME. Global variation in copy number in the human genome. Nature 2006;444(7118):444-54. 18. Mills RE, Luttig CT, Larkins CE, Beauchamp A, Tsui C, Pittard WS, Devine SE. An initial map of insertion and deletion (INDEL) variation in the human genome. Genome Res 2006;16(9):1182-90. 19. de Stahl TD, Sandgren J, Piotrowski A, Nord H, Andersson R, Menzel U, Bogdan A, Thuresson AC, Poplawski A, von Tell D, Hansson CM, Elshafie AI, Elghazali G, Imreh S, Nordenskjold M, Upadhyaya M, Komorowski J, Bruder CE, Dumanski JP. Profiling of copy number variations (CNVs) in healthy individuals from three ethnic groups using a human genome 32 K BAC-clone-based array. Human mutation 2008;29(3):398-408. 20. McCarroll SA, Kuruvilla FG, Korn JM, Cawley S, Nemesh J, Wysok- er A, Shapero MH, de Bakker PI, Maller JB, Kirby A, Elliott AL, Par- kin M, Hubbell E, Webster T, Mei R, Veitch J, Collins PJ, Handsaker R, Lincoln S, Nizzari M, Blume J, Jones KW, Rava R, Daly MJ, Ga- briel SB, Altshuler D. Integrated detection and population-genetic analysis of SNPs and copy number variation. Nat Genet 2008;40(10):1166-74. 21. Conrad DF, Pinto D, Redon R, Feuk L, Gokcumen O, Zhang Y, Aerts J, Andrews TD, Barnes C, Campbell P, Fitzgerald T, Hu M, Ihm CH, Kristiansson K, Macarthur DG, Macdonald JR, Onyiah I, Pang AW, Robson S, Stirrups K, Valsesia A, Walter K, Wei J, Tyler-Smith C, Carter NP, Lee C, Scherer SW, Hurles ME. Origins and functional impact of copy number variation in the human genome. Nature 2009;464(7289):704-12.

77 22. Itsara A, Cooper GM, Baker C, Girirajan S, Li J, Absher D, Krauss RM, Myers RM, Ridker PM, Chasman DI, Mefford H, Ying P, Nick- erson DA, Eichler EE. Population analysis of large copy number variants and hotspots of human genetic disease. Am J Hum Genet 2009;84(2):148-61. 23. Stankiewicz P, Lupski JR. Structural variation in the human genome and its role in disease. Annual review of medicine 2010;61:437-55. 24. Zhang F, Gu W, Hurles ME, Lupski JR. Copy number variation in human health, disease, and evolution. Annu Rev Genomics Hum Genet 2009;10:451-81. 25. Beckmann JS, Estivill X, Antonarakis SE. Copy number variants and genetic traits: closer to the resolution of phenotypic to genotypic variability. Nat Rev Genet 2007;8(8):639-46. 26. McCarroll SA. Extending genome-wide association studies to copy- number variation. Hum Mol Genet 2008;17(R2):R135-42. 27. Kondrashov AS. Direct estimates of human per nucleotide mutation rates at 20 loci causing Mendelian diseases. Human mutation 2003;21(1):12-27. 28. Nachman MW, Crowell SL. Estimate of the mutation rate per nucleotide in humans. Genetics 2000;156(1):297-304. 29. Lupski JR. Genomic rearrangements and sporadic disease. Nat Genet 2007;39(7 Suppl):S43-7. 30. van Ommen GJ. Frequency of new copy number variation in humans. Nat Genet 2005;37(4):333-4. 31. Gu W, Zhang F, Lupski JR. Mechanisms for human genomic rearrangements. PathoGenetics 2008;1(1):4. 32. Lee JA, Carvalho CM, Lupski JR. A DNA replication mechanism for generating nonrecurrent rearrangements associated with genomic disorders. Cell 2007;131(7):1235-47. 33. Bi W, Sapir T, Shchelochkov OA, Zhang F, Withers MA, Hunter JV, Levy T, Shinder V, Peiffer DA, Gunderson KL, Nezarati MM, Shotts VA, Amato SS, Savage SK, Harris DJ, Day-Salvatore DL, Horner M, Lu XY, Sahoo T, Yanagawa Y, Beaudet AL, Cheung SW, Martinez S, Lupski JR, Reiner O. Increased LIS1 expression affects human and mouse brain development. Nat Genet 2009;41(2):168-77. 34. Zhang F, Khajavi M, Connolly AM, Towne CF, Batish SD, Lupski JR. The DNA replication FoSTeS/MMBIR mechanism can generate genomic, genic and exonic complex rearrangements in humans. Nat Genet 2009;41(7):849-53. 35. Hastings PJ, Ira G, Lupski JR. A microhomology-mediated break- induced replication model for the origin of human copy number variation. PLoS genetics 2009;5(1):e1000327. 36. Lupski JR, Chance PF. Hereditary motor and sensory neuropathies involving altered dosage or mutation of PMP22: the CMT1A duplication andHNPPdeletion. Philadelphia: Elsevier; 2005.

78 37. Kleinjan DA, van Heyningen V. Long-range control of gene expression: emerging mechanisms and disruption in disease. Am J Hum Ge- net 2005;76(1):8-32. 38. Kurotaki N, Shen JJ, Touyama M, Kondoh T, Visser R, Ozaki T, Ni- shimoto J, Shiihara T, Uetake K, Makita Y, Harada N, Raskin S, Brown CW, Hoglund P, Okamoto N, Lupski JR. Phenotypic consequences of genetic variation at hemizygous alleles: Sotos syndrome is a contiguous gene syndrome incorporating coagulation factor twelve (FXII) deficiency. Genet Med 2005;7(7):479-83. 39. Lejeune J, Turpin R, Gautier M. [Chromosomic diagnosis of mongol- ism.]. Archives francaises de pediatrie 1959;16:962-3. 40. Pinto D, Pagnamenta AT, Klei L, Anney R, Merico D, Regan R, Con- roy J, Magalhaes TR, Correia C, Abrahams BS, Almeida J, Bacchelli E, Bader GD, Bailey AJ, Baird G, Battaglia A, Berney T, Bolshakova N, Bolte S, Bolton PF, Bourgeron T, Brennan S, Brian J, Bryson SE, Carson AR, Casallo G, Casey J, Chung BH, Cochrane L, Corsello C, Crawford EL, Crossett A, Cytrynbaum C, Dawson G, de Jonge M, Delorme R, Drmic I, Duketis E, Duque F, Estes A, Farrar P, Fernan- dez BA, Folstein SE, Fombonne E, Freitag CM, Gilbert J, Gillberg C, Glessner JT, Goldberg J, Green A, Green J, Guter SJ, Hakonarson H, Heron EA, Hill M, Holt R, Howe JL, Hughes G, Hus V, Igliozzi R, Kim C, Klauck SM, Kolevzon A, Korvatska O, Kustanovich V, La- jonchere CM, Lamb JA, Laskawiec M, Leboyer M, Le Couteur A, Leventhal BL, Lionel AC, Liu XQ, Lord C, Lotspeich L, Lund SC, Maestrini E, Mahoney W, Mantoulan C, Marshall CR, McConachie H, McDougle CJ, McGrath J, McMahon WM, Merikangas A, Migita O, Minshew NJ, Mirza GK, Munson J, Nelson SF, Noakes C, Noor A, Nygren G, Oliveira G, Papanikolaou K, Parr JR, Parrini B, Paton T, Pickles A, Pilorge M, Piven J, Ponting CP, Posey DJ, Poustka A, Poustka F, Prasad A, Ragoussis J, Renshaw K, Rickaby J, Roberts W, Roeder K, Roge B, Rutter ML, Bierut LJ, Rice JP, Salt J, Sansom K, Sato D, Segurado R, Sequeira AF, Senman L, Shah N, Sheffield VC, Soorya L, Sousa I, Stein O, Sykes N, Stoppioni V, Strawbridge C, Tancredi R, Tansey K, Thiruvahindrapduram B, Thompson AP, Thomson S, Tryfon A, Tsiantis J, Van Engeland H, Vincent JB, Volkmar F, Wallace S, Wang K, Wang Z, Wassink TH, Webber C, Weksberg R, Wing K, Wittemeyer K, Wood S, Wu J, Yaspan BL, Zu- rawiecki D, Zwaigenbaum L, Buxbaum JD, Cantor RM, Cook EH, Coon H, Cuccaro ML, Devlin B, Ennis S, Gallagher L, Geschwind DH, Gill M, Haines JL, Hallmayer J, Miller J, Monaco AP, Nurnberg- er JI, Jr., Paterson AD, Pericak-Vance MA, Schellenberg GD, Szatma- ri P, Vicente AM, Vieland VJ, Wijsman EM, Scherer SW, Sutcliffe JS, Betancur C. Functional impact of global rare copy number variation in autism spectrum disorders. Nature 2010;466(7304):368-72.

79 41. Gonzalez E, Kulkarni H, Bolivar H, Mangano A, Sanchez R, Catano G, Nibbs RJ, Freedman BI, Quinones MP, Bamshad MJ, Murthy KK, Rovin BH, Bradley W, Clark RA, Anderson SA, O'Connell R J, Agan BK, Ahuja SS, Bologna R, Sen L, Dolan MJ, Ahuja SK. The influence of CCL3L1 gene-containing segmental duplications on HIV-1/AIDS susceptibility. Science 2005;307(5714):1434-40. 42. Kuhn L, Schramm DB, Donninger S, Meddows-Taylor S, Coovadia AH, Sherman GG, Gray GE, Tiemessen CT. African infants' CCL3 gene copies influence perinatal HIV transmission in the absence of maternal nevirapine. AIDS (London, England) 2007;21(13):1753-61. 43. Manolio TA, Collins FS, Cox NJ, Goldstein DB, Hindorff LA, Hunter DJ, McCarthy MI, Ramos EM, Cardon LR, Chakravarti A, Cho JH, Guttmacher AE, Kong A, Kruglyak L, Mardis E, Rotimi CN, Slatkin M, Valle D, Whittemore AS, Boehnke M, Clark AG, Eichler EE, Gib- son G, Haines JL, Mackay TF, McCarroll SA, Visscher PM. Finding the missing heritability of complex diseases. Nature 2009;461(7265):747-53. 44. Notini AJ, Craig JM, White SJ. Copy number variation and mosaicism. Cytogenet Genome Res 2008;123(1-4):270-7. 45. Strachan T, Read A. Human Molecular Genetics 3. 3 ed. New York: Garland Publishing; 2004. 46. Bruder CE, Piotrowski A, Gijsbers AA, Andersson R, Erickson S, de Stahl TD, Menzel U, Sandgren J, von Tell D, Poplawski A, Crowley M, Crasto C, Partridge EC, Tiwari H, Allison DB, Komorowski J, van Ommen GJ, Boomsma DI, Pedersen NL, den Dunnen JT, Wirdefeldt K, Dumanski JP. Phenotypically concordant and discordant monozygotic twins display different DNA copy-number-variation profiles. Am J Hum Genet 2008;82(3):763-71. 47. Conlin LK, Thiel BD, Bonnemann CG, Medne L, Ernst LM, Zackai EH, Deardorff MA, Krantz ID, Hakonarson H, Spinner NB. Mechan- isms of mosaicism, chimerism and uniparental disomy identified by single nucleotide polymorphism array analysis. Hum Mol Genet 2010;19(7):1263-75. 48. Erickson RP. Somatic gene mutation and human disease other than cancer: An update. Mutation research 2010. 49. Di Giacomo MC, Susca FC, Resta N, Bukvic N, Vimercati A, Guanti G. Trisomy 13 mosaicism in a phenotypically normal child: description of cytogenetic and clinical findings from early pregnancy beyond 2 years of age. Am J Med Genet A 2007;143(5):518-20. 50. Venci A, Bettio D. Tetrasomy 5p mosaicism due to an additional isochromosome 5p in a man with normal phenotype. Am J Med Genet A 2009;149A(12):2889-91. 51. McAuliffe F, Winsor EJ, Chitayat D. Tetrasomy 9p mosaicism associated with a normal phenotype. Fetal diagnosis and therapy 2005;20(3):219-22.

80 52. Iwarsson E, Sahlen S, Nordgren A. Jumping translocation in a phenotypically normal male: A study of mosaicism in spermatozoa, lymphocytes, and fibroblasts. Am J Med Genet A 2009;149A(8):1706-11. 53. Ballif BC, Rorem EA, Sundin K, Lincicum M, Gaskin S, Coppinger J, Kashork CD, Shaffer LG, Bejjani BA. Detection of low-level mosaicism by array CGH in routine diagnostic specimens. Am J Med Genet A 2006;140(24):2757-67. 54. Cheung SW, Shaw CA, Scott DA, Patel A, Sahoo T, Bacino CA, Pursley A, Li J, Erickson R, Gropman AL, Miller DT, Seashore MR, Summers AM, Stankiewicz P, Chinault AC, Lupski JR, Beaudet AL, Sutton VR. Microarray-based CGH detects chromosomal mosaicism not revealed by conventional cytogenetics. Am J Med Genet A 2007;143A(15):1679-86. 55. Javierre BM, Fernandez AF, Richter J, Al-Shahrour F, Martin-Subero JI, Rodriguez-Ubreva J, Berdasco M, Fraga MF, O'Hanlon TP, Rider LG, Jacinto FV, Lopez-Longo FJ, Dopazo J, Forn M, Peinado MA, Carreno L, Sawalha AH, Harley JB, Siebert R, Esteller M, Miller FW, Ballestar E. Changes in the pattern of DNA methylation associate with twin discordance in systemic lupus erythematosus. Genome Res 2010;20(2):170-9. 56. Haque FN, Gottesman, II, Wong AH. Not really identical: epigenetic differences in monozygotic twins and implications for twin studies in psychiatry. American journal of medical genetics 2009;151C(2):136- 41. 57. Su AI, Cooke MP, Ching KA, Hakak Y, Walker JR, Wiltshire T, Orth AP, Vega RG, Sapinoso LM, Moqrich A, Patapoutian A, Hampton GM, Schultz PG, Hogenesch JB. Large-scale analysis of the human and mouse transcriptomes. Proc Natl Acad Sci U S A 2002;99(7):4465-70. 58. Su AI, Wiltshire T, Batalov S, Lapp H, Ching KA, Block D, Zhang J, Soden R, Hayakawa M, Kreiman G, Cooke MP, Walker JR, Hoge- nesch JB. A gene atlas of the mouse and human protein-encoding transcriptomes. Proc Natl Acad Sci U S A 2004;101(16):6062-7. 59. Zhang TY, Meaney MJ. Epigenetics and the environmental regulation of the genome and its function. Annual review of psychology 2009;61:439-66, C1-3. 60. Waddington CH. Preliminary Notes on the Development of the Wings in Normal and Mutant Strains of Drosophila. Proc Natl Acad Sci U S A 1939;25(7):299-307. 61. Russo VEA, Martienssen RA, Riggs AD. Epigenetic mechanisms of gene regulation. Woodbury: Cold Spring Harbor Laboratory Press; 1996. 62. Bird A. Perceptions of epigenetics. Nature 2007;447(7143):396-8. 63. Berger SL, Kouzarides T, Shiekhattar R, Shilatifard A. An operational definition of epigenetics. Genes & development 2009;23(7):781-3.

81 64. Wigler M, Levy D, Perucho M. The somatic replication of DNA methylation. Cell 1981;24(1):33-40. 65. Robertson KD. DNA methylation and human disease. Nat Rev Genet 2005;6(8):597-610. 66. Reik W, Dean W, Walter J. Epigenetic reprogramming in mammalian development. Science 2001;293(5532):1089-93. 67. Schaefer CB, Ooi SK, Bestor TH, Bourc'his D. Epigenetic decisions in mammalian germ cells. Science 2007;316(5823):398-9. 68. Bird A. DNA methylation patterns and epigenetic memory. Genes & development 2002;16(1):6-21. 69. Feinberg AP, Vogelstein B. Hypomethylation distinguishes genes of some human cancers from their normal counterparts. Nature 1983;301(5895):89-92. 70. Gama-Sosa MA, Slagel VA, Trewyn RW, Oxenhandler R, Kuo KC, Gehrke CW, Ehrlich M. The 5-methylcytosine content of DNA from human tumors. Nucleic acids research 1983;11(19):6883-94. 71. Ehrlich M, Gama-Sosa MA, Huang LH, Midgett RM, Kuo KC, McCune RA, Gehrke C. Amount and distribution of 5-methylcytosine in human DNA from different types of tissues of cells. Nucleic acids research 1982;10(8):2709-21. 72. Yoder JA, Walsh CP, Bestor TH. Cytosine methylation and the ecolo- gy of intragenomic parasites. Trends Genet 1997;13(8):335-40. 73. Antequera F, Bird A. Number of CpG islands and genes in human and mouse. Proc Natl Acad Sci U S A 1993;90(24):11995-9. 74. Suzuki MM, Bird A. DNA methylation landscapes: provocative insights from epigenomics. Nat Rev Genet 2008;9(6):465-76. 75. Doi A, Park IH, Wen B, Murakami P, Aryee MJ, Irizarry R, Herb B, Ladd-Acosta C, Rho J, Loewer S, Miller J, Schlaeger T, Daley GQ, Feinberg AP. Differential methylation of tissue- and cancer-specific CpG island shores distinguishes human induced pluripotent stem cells, embryonic stem cells and fibroblasts. Nat Genet 2009;41(12):1350-3. 76. Kornberg RD. Chromatin structure: a repeating unit of histones and DNA. Science 1974;184(139):868-71. 77. Kornberg RD, Lorch Y. Twenty-five years of the nucleosome, fundamental particle of the eukaryote chromosome. Cell 1999;98(3):285- 94. 78. Luger K, Mader AW, Richmond RK, Sargent DF, Richmond TJ. Crystal structure of the nucleosome core particle at 2.8 A resolution. Nature 1997;389(6648):251-60. 79. Henikoff S, Ahmad K. Assembly of variant histones into chromatin. Annual review of cell and developmental biology 2005;21:133-53. 80. Margueron R, Reinberg D. Chromatin structure and the inheritance of epigenetic information. Nat Rev Genet 2010;11(4):285-96. 81. Wiemer EA. The role of microRNAs in cancer: no small matter. Eur J Cancer 2007;43(10):1529-44.

82 82. Esteller M. Epigenetics in cancer. The New England journal of medicine 2008;358(11):1148-59. 83. Zhao J, Sun BK, Erwin JA, Song JJ, Lee JT. Polycomb proteins targeted by a short repeat RNA to the mouse X chromosome. Science 2008;322(5902):750-6. 84. Rinn JL, Kertesz M, Wang JK, Squazzo SL, Xu X, Brugmann SA, Goodnough LH, Helms JA, Farnham PJ, Segal E, Chang HY. Func- tional demarcation of active and silent chromatin domains in human HOX loci by noncoding RNAs. Cell 2007;129(7):1311-23. 85. Kouskouti A, Talianidis I. Histone modifications defining active genes persist after transcriptional and mitotic inactivation. The EMBO journal 2005;24(2):347-57. 86. Bernstein BE, Meissner A, Lander ES. The mammalian epigenome. Cell 2007;128(4):669-81. 87. Kouzarides T. Chromatin modifications and their function. Cell 2007;128(4):693-705. 88. Lennartsson A, Ekwall K. Histone modification patterns and epigenetic codes. Biochimica et biophysica acta 2009;1790(9):863-8. 89. Strahl BD, Allis CD. The language of covalent histone modifications. Nature 2000;403(6765):41-5. 90. Barski A, Cuddapah S, Cui K, Roh TY, Schones DE, Wang Z, Wei G, Chepelev I, Zhao K. High-resolution profiling of histone methylations in the human genome. Cell 2007;129(4):823-37. 91. Heintzman ND, Stuart RK, Hon G, Fu Y, Ching CW, Hawkins RD, Barrera LO, Van Calcar S, Qu C, Ching KA, Wang W, Weng Z, Green RD, Crawford GE, Ren B. Distinct and predictive chromatin signatures of transcriptional promoters and enhancers in the human genome. Nat Genet 2007;39(3):311-8. 92. Mikkelsen TS, Ku M, Jaffe DB, Issac B, Lieberman E, Giannoukos G, Alvarez P, Brockman W, Kim TK, Koche RP, Lee W, Mendenhall E, O'Donovan A, Presser A, Russ C, Xie X, Meissner A, Wernig M, Jae- nisch R, Nusbaum C, Lander ES, Bernstein BE. Genome-wide maps of chromatin state in pluripotent and lineage-committed cells. Nature 2007;448(7153):553-60. 93. Wang Z, Zang C, Rosenfeld JA, Schones DE, Barski A, Cuddapah S, Cui K, Roh TY, Peng W, Zhang MQ, Zhao K. Combinatorial patterns of histone acetylations and methylations in the human genome. Nat Genet 2008;40(7):897-903. 94. Kurdistani SK, Tavazoie S, Grunstein M. Mapping global histone acetylation patterns to gene expression. Cell 2004;117(6):721-33. 95. Schubeler D, MacAlpine DM, Scalzo D, Wirbelauer C, Kooperberg C, van Leeuwen F, Gottschling DE, O'Neill LP, Turner BM, Delrow J, Bell SP, Groudine M. The histone modification pattern of active genes revealed through genome-wide chromatin analysis of a higher eukaryote. Genes & development 2004;18(11):1263-71.

83 96. Lee JH, Hart SR, Skalnik DG. Histone deacetylase activity is required for embryonic stem cell differentiation. Genesis 2004;38(1):32-8. 97. De Felice L, Tatarelli C, Mascolo MG, Gregorj C, Agostini F, Fiorini R, Gelmetti V, Pascale S, Padula F, Petrucci MT, Arcese W, Nervi C. Histone deacetylase inhibitor valproic acid enhances the cytokine- induced expansion of human hematopoietic stem cells. Cancer Res 2005;65(4):1505-13. 98. Bernstein BE, Mikkelsen TS, Xie X, Kamal M, Huebert DJ, Cuff J, Fry B, Meissner A, Wernig M, Plath K, Jaenisch R, Wagschal A, Feil R, Schreiber SL, Lander ES. A bivalent chromatin structure marks key developmental genes in embryonic stem cells. Cell 2006;125(2):315- 26. 99. Pan G, Tian S, Nie J, Yang C, Ruotti V, Wei H, Jonsdottir GA, Ste- wart R, Thomson JA. Whole-genome analysis of histone H3 lysine 4 and lysine 27 methylation in human embryonic stem cells. Cell stem cell 2007;1(3):299-312. 100. Zhao XD, Han X, Chew JL, Liu J, Chiu KP, Choo A, Orlov YL, Sung WK, Shahab A, Kuznetsov VA, Bourque G, Oh S, Ruan Y, Ng HH, Wei CL. Whole-genome mapping of histone H3 Lys4 and 27 trime- thylations reveals distinct genomic compartments in human embryonic stem cells. Cell stem cell 2007;1(3):286-98. 101. Kimura H, Tada M, Nakatsuji N, Tada T. Histone code modifications on pluripotential nuclei of reprogrammed somatic cells. Molecular and cellular biology 2004;24(13):5710-20. 102. Guenther MG, Levine SS, Boyer LA, Jaenisch R, Young RA. A chromatin landmark and transcription initiation at most promoters in human cells. Cell 2007;130(1):77-88. 103. Kim J, Guermah M, McGinty RK, Lee JS, Tang Z, Milne TA, Shilati- fard A, Muir TW, Roeder RG. RAD6-Mediated transcription-coupled H2B ubiquitylation directly stimulates H3K4 methylation in human cells. Cell 2009;137(3):459-71. 104. Shilatifard A. Molecular implementation and physiological roles for histone H3 lysine 4 (H3K4) methylation. Current opinion in cell biology 2008;20(3):341-8. 105. Flanagan JF, Mi LZ, Chruszcz M, Cymborowski M, Clines KL, Kim Y, Minor W, Rastinejad F, Khorasanizadeh S. Double chromodomains cooperate to recognize the methylated histone H3 tail. Nature 2005;438(7071):1181-5. 106. Sims RJ, 3rd, Reinberg D. Histone H3 Lys 4 methylation: caught in a bind? Genes & development 2006;20(20):2779-86. 107. Guttman M, Amit I, Garber M, French C, Lin MF, Feldser D, Huarte M, Zuk O, Carey BW, Cassady JP, Cabili MN, Jaenisch R, Mikkelsen TS, Jacks T, Hacohen N, Bernstein BE, Kellis M, Regev A, Rinn JL, Lander ES. Chromatin signature reveals over a thousand highly con-

84 served large non-coding RNAs in mammals. Nature 2009;458(7235):223-7. 108. Bracken AP, Helin K. Polycomb group proteins: navigators of lineage pathways led astray in cancer. Nat Rev Cancer 2009;9(11):773-84. 109. Martinez-Garcia E, Licht JD. Deregulation of H3K27 methylation in cancer. Nat Genet 2010;42(2):100-1. 110. Garcia M. Global Cancer Facts and Figures ACS; 2007. 111. Klein CA. Parallel progression of primary tumours and metastases. Nat Rev Cancer 2009;9(4):302-12. 112. Stratton MR, Campbell PJ, Futreal PA. The cancer genome. Nature 2009;458(7239):719-24. 113. Luo J, Solimini NL, Elledge SJ. Principles of cancer therapy: oncogene and non-oncogene addiction. Cell 2009;136(5):823-37. 114. Nowell PC. The clonal evolution of tumor cell populations. Science 1976;194(4260):23-8. 115. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000;100(1):57-70. 116. Kroemer G, Pouyssegur J. Tumor cell metabolism: cancer's Achilles' heel. Cancer Cell 2008;13(6):472-82. 117. Talbot SJ, Crawford DH. Viruses and tumours--an update. Eur J Can- cer 2004;40(13):1998-2005. 118. Hartwell LH, Kastan MB. Cell cycle control and cancer. Science 1994;266(5192):1821-8. 119. Maser RS, DePinho RA. Connecting chromosomes, crisis, and cancer. Science 2002;297(5581):565-9. 120. Bartkova J, Horejsi Z, Koed K, Kramer A, Tort F, Zieger K, Guldberg P, Sehested M, Nesland JM, Lukas C, Orntoft T, Lukas J, Bartek J. DNA damage response as a candidate anti-cancer barrier in early human tumorigenesis. Nature 2005;434(7035):864-70. 121. Harper JW, Elledge SJ. The DNA damage response: ten years after. Molecular cell 2007;28(5):739-45. 122. Denoyelle C, Abou-Rjaily G, Bezrookove V, Verhaegen M, Johnson TM, Fullen DR, Pointer JN, Gruber SB, Su LD, Nikiforov MA, Kaufman RJ, Bastian BC, Soengas MS. Anti-oncogenic role of the endoplasmic reticulum differentially activated by mutations in the MAPK pathway. Nature cell biology 2006;8(10):1053-63. 123. Komarova NL, Lengauer C, Vogelstein B, Nowak MA. Dynamics of genetic instability in sporadic and familial colorectal cancer. Cancer biology & therapy 2002;1(6):685-92. 124. Ganem NJ, Godinho SA, Pellman D. A mechanism linking extra cen- trosomes to chromosomal instability. Nature 2009;460(7252):278-82. 125. Negrini S, Gorgoulis VG, Halazonetis TD. Genomic instability--an evolving hallmark of cancer. Nature reviews 2010;11(3):220-8. 126. Warburg O. On the origin of cancer cells. Science 1956;123(3191):309-14.

85 127. Gogvadze V, Orrenius S, Zhivotovsky B. Mitochondria in cancer cells: what is so special about them? Trends in cell biology 2008;18(4):165-73. 128. von Hansemann D. Ueber asymmetrische Zelltheilung in epithel Krebsen und deren biologische Bedeutung: Virchows Arch. Path. Anat; 1890. 129. Boveri T. Zur Frage der Entstehung Maligner Tumoren; 1914. 130. Avery OT, Macleod CM, McCarty M. Studies on the Chemical Nature of the Substance Inducing Transformation of Pneumococcal Types : Induction of Transformation by a Desoxyribonucleic Acid Fraction Isolated from Pneumococcus Type Iii. The Journal of experimental medicine 1944;79(2):137-58. 131. Loeb LA, Harris CC. Advances in chemical carcinogenesis: a histori- cal review and prospective. Cancer Res 2008;68(17):6863-72. 132. Rowley JD. Letter: A new consistent chromosomal abnormality in chronic myelogenous leukaemia identified by quinacrine fluorescence and Giemsa staining. Nature 1973;243(5405):290-3. 133. Reddy EP, Reynolds RK, Santos E, Barbacid M. A point mutation is responsible for the acquisition of transforming properties by the T24 human bladder carcinoma oncogene. Nature 1982;300(5888):149-52. 134. Knudson AG, Jr. Retinoblastoma: a prototypic hereditary neoplasm. Seminars in oncology 1978;5(1):57-60. 135. Chari R, Thu KL, Wilson IM, Lockwood WW, Lonergan KM, Coe BP, Malloff CA, Gazdar AF, Lam S, Garnis C, MacAulay CE, Alva- rez CE, Lam WL. Integrating the multiple dimensions of genomic and epigenomic landscapes of cancer. Cancer metastasis reviews 2010;29(1):73-93. 136. Degenhardt YY, Wooster R, McCombie RW, Lucito R, Powers S. High-content analysis of cancer genome DNA alterations. Current opinion in genetics & development 2008;18(1):68-72. 137. Pollack JR, Sorlie T, Perou CM, Rees CA, Jeffrey SS, Lonning PE, Tibshirani R, Botstein D, Borresen-Dale AL, Brown PO. Microarray analysis reveals a major direct role of DNA copy number alteration in the transcriptional program of human breast tumors. Proc Natl Acad Sci U S A 2002;99(20):12963-8. 138. Torres EM, Sokolsky T, Tucker CM, Chan LY, Boselli M, Dunham MJ, Amon A. Effects of aneuploidy on cellular physiology and cell division in haploid yeast. Science 2007;317(5840):916-24. 139. Papp B, Pal C, Hurst LD. Dosage sensitivity and the evolution of gene families in yeast. Nature 2003;424(6945):194-7. 140. Albertson DG. Gene amplification in cancer. Trends Genet 2006;22(8):447-55. 141. Santarius T, Shipley J, Brewer D, Stratton MR, Cooper CS. A census of amplified and overexpressed human cancer genes. Nat Rev Cancer 2010;10(1):59-64.

86 142. Nevins JR, Potti A. Mining gene expression profiles: expression signatures as cancer phenotypes. Nat Rev Genet 2007;8(8):601-9. 143. Prensner JR, Chinnaiyan AM. Oncogenic gene fusions in epithelial carcinomas. Current opinion in genetics & development 2009;19(1):82-91. 144. Golub TR, Slonim DK, Tamayo P, Huard C, Gaasenbeek M, Mesirov JP, Coller H, Loh ML, Downing JR, Caligiuri MA, Bloomfield CD, Lander ES. Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 1999;286(5439):531-7. 145. Garraway LA, Widlund HR, Rubin MA, Getz G, Berger AJ, Ramas- wamy S, Beroukhim R, Milner DA, Granter SR, Du J, Lee C, Wagner SN, Li C, Golub TR, Rimm DL, Meyerson ML, Fisher DE, Sellers WR. Integrative genomic analyses identify MITF as a lineage survival oncogene amplified in malignant melanoma. Nature 2005;436(7047):117-22. 146. Beroukhim R, Mermel CH, Porter D, Wei G, Raychaudhuri S, Dono- van J, Barretina J, Boehm JS, Dobson J, Urashima M, Mc Henry KT, Pinchback RM, Ligon AH, Cho YJ, Haery L, Greulich H, Reich M, Winckler W, Lawrence MS, Weir BA, Tanaka KE, Chiang DY, Bass AJ, Loo A, Hoffman C, Prensner J, Liefeld T, Gao Q, Yecies D, Sig- noretti S, Maher E, Kaye FJ, Sasaki H, Tepper JE, Fletcher JA, Taber- nero J, Baselga J, Tsao MS, Demichelis F, Rubin MA, Janne PA, Daly MJ, Nucera C, Levine RL, Ebert BL, Gabriel S, Rustgi AK, Antones- cu CR, Ladanyi M, Letai A, Garraway LA, Loda M, Beer DG, True LD, Okamoto A, Pomeroy SL, Singer S, Golub TR, Lander ES, Getz G, Sellers WR, Meyerson M. The landscape of somatic copy-number alteration across human cancers. Nature 2010;463(7283):899-905. 147. Fraga MF, Herranz M, Espada J, Ballestar E, Paz MF, Ropero S, Er- kek E, Bozdogan O, Peinado H, Niveleau A, Mao JH, Balmain A, Ca- no A, Esteller M. A mouse skin multistage carcinogenesis model re- flects the aberrant DNA methylation patterns of human tumors. Can- cer Res 2004;64(16):5527-34. 148. Cui H, Cruz-Correa M, Giardiello FM, Hutcheon DF, Kafonek DR, Brandenburg S, Wu Y, He X, Powe NR, Feinberg AP. Loss of IGF2 imprinting: a potential marker of colorectal cancer risk. Science 2003;299(5613):1753-5. 149. Kaneda A, Feinberg AP. Loss of imprinting of IGF2: a common epigenetic modifier of intestinal tumor risk. Cancer Res 2005;65(24):11236-40. 150. Hirst M, Marra MA. Epigenetics and human disease. The international journal of biochemistry & cell biology 2009;41(1):136-46. 151. Costello JF, Fruhwald MC, Smiraglia DJ, Rush LJ, Robertson GP, Gao X, Wright FA, Feramisco JD, Peltomaki P, Lang JC, Schuller DE, Yu L, Bloomfield CD, Caligiuri MA, Yates A, Nishikawa R, Su

87 Huang H, Petrelli NJ, Zhang X, O'Dorisio MS, Held WA, Cavenee WK, Plass C. Aberrant CpG-island methylation has non-random and tumour-type-specific patterns. Nat Genet 2000;24(2):132-8. 152. Esteller M, Corn PG, Baylin SB, Herman JG. A gene hypermethylation profile of human cancer. Cancer Res 2001;61(8):3225-9. 153. Shilatifard A. Chromatin modifications by methylation and ubiquitina- tion: implications in the regulation of gene expression. Annual review of biochemistry 2006;75:243-69. 154. Schneider R, Bannister AJ, Kouzarides T. Unsafe SETs: histone lysine methyltransferases and cancer. Trends in biochemical sciences 2002;27(8):396-402. 155. Kahl P, Gullotti L, Heukamp LC, Wolf S, Friedrichs N, Vorreuther R, Solleder G, Bastian PJ, Ellinger J, Metzger E, Schule R, Buettner R. Androgen receptor coactivators lysine-specific histone demethylase 1 and four and a half LIM domain protein 2 predict risk of prostate cancer recurrence. Cancer Res 2006;66(23):11341-7. 156. Barrett A, Madsen B, Copier J, Lu PJ, Cooper L, Scibetta AG, Bur- chell J, Taylor-Papadimitriou J. PLU-1 nuclear protein, which is upregulated in breast cancer, shows restricted expression in normal human adult tissues: a new cancer/testis antigen? Int J Cancer 2002;101(6):581-8. 157. van Haaften G, Dalgliesh GL, Davies H, Chen L, Bignell G, Green- man C, Edkins S, Hardy C, O'Meara S, Teague J, Butler A, Hinton J, Latimer C, Andrews J, Barthorpe S, Beare D, Buck G, Campbell PJ, Cole J, Forbes S, Jia M, Jones D, Kok CY, Leroy C, Lin ML, McBride DJ, Maddison M, Maquire S, McLay K, Menzies A, Miro- nenko T, Mulderrig L, Mudie L, Pleasance E, Shepherd R, Smith R, Stebbings L, Stephens P, Tang G, Tarpey PS, Turner R, Turrell K, Va- rian J, West S, Widaa S, Wray P, Collins VP, Ichimura K, Law S, Wong J, Yuen ST, Leung SY, Tonon G, DePinho RA, Tai YT, Ander- son KC, Kahnoski RJ, Massie A, Khoo SK, Teh BT, Stratton MR, Fu- treal PA. Somatic mutations of the histone H3K27 demethylase gene UTX in human cancer. Nat Genet 2009;41(5):521-3. 158. Yang ZQ, Imoto I, Fukuda Y, Pimkhaokham A, Shimada Y, Imamura M, Sugano S, Nakamura Y, Inazawa J. Identification of a novel gene, GASC1, within an amplicon at 9p23-24 frequently detected in eso- phageal cancer cell lines. Cancer Res 2000;60(17):4735-9. 159. Neff T, Armstrong SA. Chromatin maps, histone modifications and leukemia. Leukemia 2009;23(7):1243-51. 160. Ke XS, Qu Y, Rostad K, Li WC, Lin B, Halvorsen OJ, Haukaas SA, Jonassen I, Petersen K, Goldfinger N, Rotter V, Akslen LA, Oyan AM, Kalland KH. Genome-wide profiling of histone h3 lysine 4 and lysine 27 trimethylation reveals an epigenetic signature in prostate carcinogenesis. PLoS ONE 2009;4(3):e4687.

88 161. Hamamoto R, Furukawa Y, Morita M, Iimura Y, Silva FP, Li M, Yagyu R, Nakamura Y. SMYD3 encodes a histone methyltransferase involved in the proliferation of cancer cells. Nature cell biology 2004;6(8):731-40. 162. Simon JA, Lange CA. Roles of the EZH2 histone methyltransferase in cancer epigenetics. Mutation research 2008;647(1-2):21-9. 163. Vire E, Brenner C, Deplus R, Blanchon L, Fraga M, Didelot C, Morey L, Van Eynde A, Bernard D, Vanderwinden JM, Bollen M, Esteller M, Di Croce L, de Launoit Y, Fuks F. The Polycomb group protein EZH2 directly controls DNA methylation. Nature 2006;439(7078):871-4. 164. Widschwendter M, Fiegl H, Egle D, Mueller-Holzner E, Spizzo G, Marth C, Weisenberger DJ, Campan M, Young J, Jacobs I, Laird PW. Epigenetic stem cell signature in cancer. Nat Genet 2007;39(2):157-8. 165. Schlesinger Y, Straussman R, Keshet I, Farkash S, Hecht M, Zim- merman J, Eden E, Yakhini Z, Ben-Shushan E, Reubinoff BE, Berg- man Y, Simon I, Cedar H. Polycomb-mediated methylation on Lys27 of histone H3 pre-marks genes for de novo methylation in cancer. Nat Genet 2007;39(2):232-6. 166. Kondo Y, Shen L, Cheng AS, Ahmed S, Boumber Y, Charo C, Ya- mochi T, Urano T, Furukawa K, Kwabi-Addo B, Gold DL, Sekido Y, Huang TH, Issa JP. Gene silencing in cancer by histone H3 lysine 27 trimethylation independent of promoter DNA methylation. Nat Genet 2008;40(6):741-50. 167. Gibbons RJ. Histone modifying and chromatin remodelling enzymes in cancer and dysplastic syndromes. Hum Mol Genet 2005;14 Spec No 1:R85-92. 168. Fraga MF, Ballestar E, Villar-Garea A, Boix-Chornet M, Espada J, Schotta G, Bonaldi T, Haydon C, Ropero S, Petrie K, Iyer NG, Perez- Rosado A, Calvo E, Lopez JA, Cano A, Calasanz MJ, Colomer D, Pi- ris MA, Ahn N, Imhof A, Caldas C, Jenuwein T, Esteller M. Loss of acetylation at Lys16 and trimethylation at Lys20 of histone H4 is a common hallmark of human cancer. Nat Genet 2005;37(4):391-400. 169. Seligson DB, Horvath S, Shi T, Yu H, Tze S, Grunstein M, Kurdistani SK. Global histone modification patterns predict risk of prostate cancer recurrence. Nature 2005;435(7046):1262-6. 170. Yu J, Yu J, Rhodes DR, Tomlins SA, Cao X, Chen G, Mehra R, Wang X, Ghosh D, Shah RB, Varambally S, Pienta KJ, Chinnaiyan AM. A polycomb repression signature in metastatic prostate cancer predicts cancer outcome. Cancer Res 2007;67(22):10657-63. 171. Nguyen CT, Weisenberger DJ, Velicescu M, Gonzales FA, Lin JC, Liang G, Jones PA. Histone H3-lysine 9 methylation is associated with aberrant gene silencing in cancer cells and is rapidly reversed by 5-aza-2'-deoxycytidine. Cancer Res 2002;62(22):6456-61.

89 172. Sjoblom T, Jones S, Wood LD, Parsons DW, Lin J, Barber TD, Man- delker D, Leary RJ, Ptak J, Silliman N, Szabo S, Buckhaults P, Farrell C, Meeh P, Markowitz SD, Willis J, Dawson D, Willson JK, Gazdar AF, Hartigan J, Wu L, Liu C, Parmigiani G, Park BH, Bachman KE, Papadopoulos N, Vogelstein B, Kinzler KW, Velculescu VE. The consensus coding sequences of human breast and colorectal cancers. Science 2006;314(5797):268-74. 173. Jones S, Zhang X, Parsons DW, Lin JC, Leary RJ, Angenendt P, Mankoo P, Carter H, Kamiyama H, Jimeno A, Hong SM, Fu B, Lin MT, Calhoun ES, Kamiyama M, Walter K, Nikolskaya T, Nikolsky Y, Hartigan J, Smith DR, Hidalgo M, Leach SD, Klein AP, Jaffee EM, Goggins M, Maitra A, Iacobuzio-Donahue C, Eshleman JR, Kern SE, Hruban RH, Karchin R, Papadopoulos N, Parmigiani G, Vogelstein B, Velculescu VE, Kinzler KW. Core signaling pathways in human pan- creatic cancers revealed by global genomic analyses. Science 2008;321(5897):1801-6. 174. Pleasance ED, Stephens PJ, O'Meara S, McBride DJ, Meynert A, Jones D, Lin ML, Beare D, Lau KW, Greenman C, Varela I, Nik- Zainal S, Davies HR, Ordonez GR, Mudie LJ, Latimer C, Edkins S, Stebbings L, Chen L, Jia M, Leroy C, Marshall J, Menzies A, Butler A, Teague JW, Mangion J, Sun YA, McLaughlin SF, Peckham HE, Tsung EF, Costa GL, Lee CC, Minna JD, Gazdar A, Birney E, Rhodes MD, McKernan KJ, Stratton MR, Futreal PA, Campbell PJ. A small- cell lung cancer genome with complex signatures of tobacco exposure. Nature 2010;463(7278):184-90. 175. Ding L, Ellis MJ, Li S, Larson DE, Chen K, Wallis JW, Harris CC, McLellan MD, Fulton RS, Fulton LL, Abbott RM, Hoog J, Dooling DJ, Koboldt DC, Schmidt H, Kalicki J, Zhang Q, Chen L, Lin L, Wendl MC, McMichael JF, Magrini VJ, Cook L, McGrath SD, Vick- ery TL, Appelbaum E, Deschryver K, Davies S, Guintoli T, Lin L, Crowder R, Tao Y, Snider JE, Smith SM, Dukes AF, Sanderson GE, Pohl CS, Delehaunty KD, Fronick CC, Pape KA, Reed JS, Robinson JS, Hodges JS, Schierding W, Dees ND, Shen D, Locke DP, Wiechert ME, Eldred JM, Peck JB, Oberkfell BJ, Lolofie JT, Du F, Hawkins AE, O'Laughlin MD, Bernard KE, Cunningham M, Elliott G, Mason MD, Thompson DM, Jr., Ivanovich JL, Goodfellow PJ, Perou CM, Weinstock GM, Aft R, Watson M, Ley TJ, Wilson RK, Mardis ER. Genome remodelling in a basal-like breast cancer metastasis and xe- nograft. Nature 2010;464(7291):999-1005. 176. Greenman C, Stephens P, Smith R, Dalgliesh GL, Hunter C, Bignell G, Davies H, Teague J, Butler A, Stevens C, Edkins S, O'Meara S, Vastrik I, Schmidt EE, Avis T, Barthorpe S, Bhamra G, Buck G, Choudhury B, Clements J, Cole J, Dicks E, Forbes S, Gray K, Halli- day K, Harrison R, Hills K, Hinton J, Jenkinson A, Jones D, Menzies A, Mironenko T, Perry J, Raine K, Richardson D, Shepherd R, Small

90 A, Tofts C, Varian J, Webb T, West S, Widaa S, Yates A, Cahill DP, Louis DN, Goldstraw P, Nicholson AG, Brasseur F, Looijenga L, Weber BL, Chiew YE, DeFazio A, Greaves MF, Green AR, Campbell P, Birney E, Easton DF, Chenevix-Trench G, Tan MH, Khoo SK, Teh BT, Yuen ST, Leung SY, Wooster R, Futreal PA, Stratton MR. Pat- terns of somatic mutation in human cancer genomes. Nature 2007;446(7132):153-8. 177. Parsons DW, Jones S, Zhang X, Lin JC, Leary RJ, Angenendt P, Mankoo P, Carter H, Siu IM, Gallia GL, Olivi A, McLendon R, Ra- sheed BA, Keir S, Nikolskaya T, Nikolsky Y, Busam DA, Tekleab H, Diaz LA, Jr., Hartigan J, Smith DR, Strausberg RL, Marie SK, Shinjo SM, Yan H, Riggins GJ, Bigner DD, Karchin R, Papadopoulos N, Parmigiani G, Vogelstein B, Velculescu VE, Kinzler KW. An integrated genomic analysis of human glioblastoma multiforme. Science 2008;321(5897):1807-12. 178. Wood LD, Parsons DW, Jones S, Lin J, Sjoblom T, Leary RJ, Shen D, Boca SM, Barber T, Ptak J, Silliman N, Szabo S, Dezso Z, Us- tyanksky V, Nikolskaya T, Nikolsky Y, Karchin R, Wilson PA, Ka- minker JS, Zhang Z, Croshaw R, Willis J, Dawson D, Shipitsin M, Willson JK, Sukumar S, Polyak K, Park BH, Pethiyagoda CL, Pant PV, Ballinger DG, Sparks AB, Hartigan J, Smith DR, Suh E, Papado- poulos N, Buckhaults P, Markowitz SD, Parmigiani G, Kinzler KW, Velculescu VE, Vogelstein B. The genomic landscapes of human breast and colorectal cancers. Science 2007;318(5853):1108-13. 179. Network CGAR. Comprehensive genomic characterization defines human glioblastoma genes and core pathways. Nature 2008;455(7216):1061-8. 180. Ding L, Getz G, Wheeler DA, Mardis ER, McLellan MD, Cibulskis K, Sougnez C, Greulich H, Muzny DM, Morgan MB, Fulton L, Fulton RS, Zhang Q, Wendl MC, Lawrence MS, Larson DE, Chen K, Dool- ing DJ, Sabo A, Hawes AC, Shen H, Jhangiani SN, Lewis LR, Hall O, Zhu Y, Mathew T, Ren Y, Yao J, Scherer SE, Clerc K, Metcalf GA, Ng B, Milosavljevic A, Gonzalez-Garay ML, Osborne JR, Meyer R, Shi X, Tang Y, Koboldt DC, Lin L, Abbott R, Miner TL, Pohl C, Fe- well G, Haipek C, Schmidt H, Dunford-Shore BH, Kraja A, Crosby SD, Sawyer CS, Vickery T, Sander S, Robinson J, Winckler W, Baldwin J, Chirieac LR, Dutt A, Fennell T, Hanna M, Johnson BE, Onofrio RC, Thomas RK, Tonon G, Weir BA, Zhao X, Ziaugra L, Zody MC, Giordano T, Orringer MB, Roth JA, Spitz MR, Wistuba, II, Ozenberger B, Good PJ, Chang AC, Beer DG, Watson MA, Ladanyi M, Broderick S, Yoshizawa A, Travis WD, Pao W, Province MA, Weinstock GM, Varmus HE, Gabriel SB, Lander ES, Gibbs RA, Meyerson M, Wilson RK. Somatic mutations affect key pathways in lung adenocarcinoma. Nature 2008;455(7216):1069-75.

91 181. Samuels Y, Wang Z, Bardelli A, Silliman N, Ptak J, Szabo S, Yan H, Gazdar A, Powell SM, Riggins GJ, Willson JK, Markowitz S, Kinzler KW, Vogelstein B, Velculescu VE. High frequency of mutations of the PIK3CA gene in human cancers. Science 2004;304(5670):554. 182. Davies H, Bignell GR, Cox C, Stephens P, Edkins S, Clegg S, Teague J, Woffendin H, Garnett MJ, Bottomley W, Davis N, Dicks E, Ewing R, Floyd Y, Gray K, Hall S, Hawes R, Hughes J, Kosmidou V, Men- zies A, Mould C, Parker A, Stevens C, Watt S, Hooper S, Wilson R, Jayatilake H, Gusterson BA, Cooper C, Shipley J, Hargrave D, Prit- chard-Jones K, Maitland N, Chenevix-Trench G, Riggins GJ, Bigner DD, Palmieri G, Cossu A, Flanagan A, Nicholson A, Ho JW, Leung SY, Yuen ST, Weber BL, Seigler HF, Darrow TL, Paterson H, Marais R, Marshall CJ, Wooster R, Stratton MR, Futreal PA. Mutations of the BRAF gene in human cancer. Nature 2002;417(6892):949-54. 183. Dalgliesh GL, Furge K, Greenman C, Chen L, Bignell G, Butler A, Davies H, Edkins S, Hardy C, Latimer C, Teague J, Andrews J, Bar- thorpe S, Beare D, Buck G, Campbell PJ, Forbes S, Jia M, Jones D, Knott H, Kok CY, Lau KW, Leroy C, Lin ML, McBride DJ, Maddi- son M, Maguire S, McLay K, Menzies A, Mironenko T, Mulderrig L, Mudie L, O'Meara S, Pleasance E, Rajasingham A, Shepherd R, Smith R, Stebbings L, Stephens P, Tang G, Tarpey PS, Turrell K, Dykema KJ, Khoo SK, Petillo D, Wondergem B, Anema J, Kahnoski RJ, Teh BT, Stratton MR, Futreal PA. Systematic sequencing of renal carcinoma reveals inactivation of histone modifying genes. Nature 2010;463(7279):360-3. 184. Stephens PJ, McBride DJ, Lin ML, Varela I, Pleasance ED, Simpson JT, Stebbings LA, Leroy C, Edkins S, Mudie LJ, Greenman CD, Jia M, Latimer C, Teague JW, Lau KW, Burton J, Quail MA, Swerdlow H, Churcher C, Natrajan R, Sieuwerts AM, Martens JW, Silver DP, Langerod A, Russnes HE, Foekens JA, Reis-Filho JS, van 't Veer L, Richardson AL, Borresen-Dale AL, Campbell PJ, Futreal PA, Stratton MR. Complex landscapes of somatic rearrangement in human breast cancer genomes. Nature 2009;462(7276):1005-10. 185. ICGC. International network of cancer genome projects. Nature 2010;464(7291):993-8. 186. Chin L, Gray JW. Translating insights from the cancer genome into clinical practice. Nature 2008;452(7187):553-63. 187. Slamon DJ, Leyland-Jones B, Shak S, Fuchs H, Paton V, Bajamonde A, Fleming T, Eiermann W, Wolter J, Pegram M, Baselga J, Norton L. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. The New England journal of medicine 2001;344(11):783-92. 188. Druker BJ, Talpaz M, Resta DJ, Peng B, Buchdunger E, Ford JM, Lydon NB, Kantarjian H, Capdeville R, Ohno-Jones S, Sawyers CL. Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine

92 kinase in chronic myeloid leukemia. The New England journal of medicine 2001;344(14):1031-7. 189. Fränkel. Classics in oncology. A case of bilateral completely latent adrenal tumor and concurrent nephritis with changes in the circulatory system and retinitis: Felix Frankel, 1886. CA: a cancer journal for clinicians 1984;34(2):93-106. 190. Petri BJ, van Eijck CH, de Herder WW, Wagner A, de Krijger RR. Phaeochromocytomas and sympathetic paragangliomas. The British journal of surgery 2009;96(12):1381-92. 191. Karagiannis A, Mikhailidis DP, Athyros VG, Harsoulis F. Pheochro- mocytoma: an update on genetics and management. Endocr Relat Cancer 2007;14(4):935-56. 192. Adler JT, Meyer-Rochow GY, Chen H, Benn DE, Robinson BG, Sip- pel RS, Sidhu SB. Pheochromocytoma: current approaches and future directions. Oncologist 2008;13(7):779-93. 193. Beard CM, Sheps SG, Kurland LT, Carney JA, Lie JT. Occurrence of pheochromocytoma in Rochester, Minnesota, 1950 through 1979. Mayo Clinic proceedings 1983;58(12):802-4. 194. Stenstrom G, Svardsudd K. Pheochromocytoma in Sweden 1958- 1981. An analysis of the National Cancer Registry Data. Acta medica Scandinavica 1986;220(3):225-32. 195. Santarpia L, Habra MA, Jimenez C. Malignant Pheochromocytomas and Paragangliomas: Molecular Signaling Pathways and Emerging Therapies. Horm Metab Res 2009. 196. Amar L, Servais A, Gimenez-Roqueplo AP, Zinzindohoue F, Chatel- lier G, Plouin PF. Year of diagnosis, features at presentation, and risk of recurrence in patients with pheochromocytoma or secreting paraganglioma. The Journal of clinical endocrinology and metabolism 2005;90(4):2110-6. 197. Chrisoulidou A, Kaltsas G, Ilias I, Grossman AB. The diagnosis and management of malignant phaeochromocytoma and paraganglioma. Endocr Relat Cancer 2007;14(3):569-85. 198. Elder EE, Elder G, Larsson C. Pheochromocytoma and functional paraganglioma syndrome: No longer the 10% tumor. J Surg Oncol 2005;89(3):193-201. 199. Amar L, Bertherat J, Baudin E, Ajzenberg C, Bressac-de Paillerets B, Chabre O, Chamontin B, Delemer B, Giraud S, Murat A, Niccoli-Sire P, Richard S, Rohmer V, Sadoul JL, Strompf L, Schlumberger M, Bertagna X, Plouin PF, Jeunemaitre X, Gimenez-Roqueplo AP. Ge- netic testing in pheochromocytoma or functional paraganglioma. J Clin Oncol 2005;23(34):8812-8. 200. Gimenez-Roqueplo AP, Favier J, Rustin P, Rieubland C, Crespin M, Nau V, Khau Van Kien P, Corvol P, Plouin PF, Jeunemaitre X. Muta- tions in the SDHB gene are associated with extra-adrenal and/or malignant phaeochromocytomas. Cancer Res 2003;63(17):5615-21.

93 201. DeLellis RA, Lloyd RV, Heitz PU, Eng C. WHO Classification of Tumours—Pathology and Genetics of Tumours of Endocrine Organs. Lyon: IARC Press; 2004. 202. Plouin PF, Chatellier G, Fofol I, Corvol P. Tumor recurrence and hypertension persistence after successful pheochromocytoma operation. Hypertension 1997;29(5):1133-9. 203. Wu D, Tischler AS, Lloyd RV, DeLellis RA, de Krijger R, van Ne- derveen F, Nose V. Observer variation in the application of the Pheochromocytoma of the Adrenal Gland Scaled Score. The American journal of surgical pathology 2009;33(4):599-608. 204. Suh I, Shibru D, Eisenhofer G, Pacak K, Duh QY, Clark OH, Kebe- bew E. Candidate Genes Associated With Malignant Pheochromocy- tomas by Genome-Wide Expression Profiling. Annals of surgery 2009. 205. Anouar Y, Yon L, Guillemot J, Thouennon E, Barbier L, Gimenez- Roqueplo AP, Bertherat J, Lefebvre H, Klein M, Muresan M, Grouz- mann E, Plouin PF, Vaudry H, Elkahloun AG. Development of novel tools for the diagnosis and prognosis of pheochromocytoma using peptide marker immunoassay and gene expression profiling approaches. Annals of the New York Academy of Sciences 2006;1073:533-40. 206. Brouwers FM, Elkahloun AG, Munson PJ, Eisenhofer G, Barb J, Li- nehan WM, Lenders JW, De Krijger R, Mannelli M, Udelsman R, Oc- al IT, Shulkin BL, Bornstein SR, Breza J, Ksinantova L, Pacak K. Gene expression profiling of benign and malignant pheochromocytoma. Annals of the New York Academy of Sciences 2006;1073:541-56. 207. Thouennon E, Elkahloun AG, Guillemot J, Gimenez-Roqueplo AP, Bertherat J, Pierre A, Ghzili H, Grumolato L, Muresan M, Klein M, Lefebvre H, Ouafik L, Vaudry H, Plouin PF, Yon L, Anouar Y. Iden- tification of potential gene markers and insights into the pathophysiol- ogy of pheochromocytoma malignancy. The Journal of clinical endocrinology and metabolism 2007;92(12):4865-72. 208. Bjorklund P, Cupisti K, Fryknas M, Isaksson A, Willenberg HS, Akerstrom G, Hellman P, Westin G. Stathmin as a marker for malignancy in pheochromocytomas. Exp Clin Endocrinol Diabetes 2009;118(1):27-30. 209. Yuan W, Wang W, Cui B, Su T, Ge Y, Jiang L, Zhou W, Ning G. Overexpression of ERBB-2 was more frequently detected in malignant than benign pheochromocytomas by multiplex ligation-dependent probe amplification and immunohistochemistry. Endocr Relat Cancer 2008;15(1):343-50. 210. Waldmann J, Slater EP, Langer P, Buchholz M, Ramaswamy A, Walz MK, Schmid KW, Feldmann G, Bartsch DK, Fendrich V. Expression of the transcription factor snail and its target gene twist are associated with malignancy in pheochromocytomas. Annals of surgical oncology 2009;16(7):1997-2005.

94 211. Hayry V, Salmenkivi K, Arola J, Heikkila P, Haglund C, Sariola H. High frequency of SNAIL-expressing cells confirms and predicts metastatic potential of phaeochromocytoma. Endocr Relat Cancer 2009;16(4):1211-8. 212. Ohta S, Lai EW, Pang AL, Brouwers FM, Chan WY, Eisenhofer G, de Krijger R, Ksinantova L, Breza J, Blazicek P, Kvetnansky R, Wesley RA, Pacak K. Downregulation of metastasis suppressor genes in malignant pheochromocytoma. Int J Cancer 2005;114(1):139-43. 213. van Nederveen FH, Perren A, Dannenberg H, Petri BJ, Dinjens WN, Komminoth P, de Krijger RR. PTEN gene loss, but not mutation, in benign and malignant phaeochromocytomas. The Journal of pathology 2006;209(2):274-80. 214. Korpershoek E, Loonen AJ, Corvers S, van Nederveen FH, Jonkers J, Ma X, Ziel-van der Made A, Korsten H, Trapman J, Dinjens WN, de Krijger RR. Conditional Pten knock-out mice: a model for metastatic phaeochromocytoma. The Journal of pathology 2009;217(4):597-604. 215. Edstrom E, Mahlamaki E, Nord B, Kjellman M, Karhu R, Hoog A, Goncharov N, Teh BT, Backdahl M, Larsson C. Comparative genomic hybridization reveals frequent losses of chromosomes 1p and 3q in pheochromocytomas and abdominal paragangliomas, suggesting a common genetic etiology. Am J Pathol 2000;156(2):651-9. 216. Cascon A, Ruiz-Llorente S, Rodriguez-Perales S, Honrado E, Marti- nez-Ramirez A, Leton R, Montero-Conde C, Benitez J, Dopazo J, Ci- gudosa JC, Robledo M. A novel candidate region linked to development of both pheochromocytoma and head/neck paraganglioma. Genes Chromosomes Cancer 2005;42(3):260-8. 217. Dannenberg H, Speel EJ, Zhao J, Saremaslani P, van Der Harst E, Roth J, Heitz PU, Bonjer HJ, Dinjens WN, Mooi WJ, Komminoth P, de Krijger RR. Losses of chromosomes 1p and 3q are early genetic events in the development of sporadic pheochromocytomas. Am J Pa- thol 2000;157(2):353-9. 218. August C, August K, Schroeder S, Bahn H, Hinze R, Baba HA, Kerst- ing C, Buerger H. CGH and CD 44/MIB-1 immunohistochemistry are helpful to distinguish metastasized from nonmetastasized sporadic pheochromocytomas. Mod Pathol 2004;17(9):1119-28. 219. Gimenez-Roqueplo AP, Burnichon N, Amar L, Favier J, Jeunemaitre X, Plouin PF. Recent advances in the genetics of phaeochromocytoma and functional paraganglioma. Clinical and experimental pharmacol- ogy & physiology 2008;35(4):376-9. 220. Bornstein SR, Gimenez-Roqueplo AP. Genetic testing in pheochromocytoma: increasing importance for clinical decision making. Annals of the New York Academy of Sciences 2006;1073:94-103. 221. Bausch B, Boedeker CC, Berlis A, Brink I, Cybulla M, Walz MK, Januszewicz A, Letizia C, Opocher G, Eng C, Neumann HP. Genetic and clinical investigation of pheochromocytoma: a 22-year expe-

95 rience, from Freiburg, Germany to international effort. Annals of the New York Academy of Sciences 2006;1073:122-37. 222. Latif F, Tory K, Gnarra J, Yao M, Duh FM, Orcutt ML, Stackhouse T, Kuzmin I, Modi W, Geil L, et al. Identification of the von Hippel- Lindau disease tumor suppressor gene. Science 1993;260(5112):1317- 20. 223. Mulligan LM, Kwok JB, Healey CS, Elsdon MJ, Eng C, Gardner E, Love DR, Mole SE, Moore JK, Papi L, et al. Germ-line mutations of the RET proto-oncogene in multiple endocrine neoplasia type 2A. Na- ture 1993;363(6428):458-60. 224. Xu W, Mulligan LM, Ponder MA, Liu L, Smith BA, Mathew CG, Ponder BA. Loss of NF1 alleles in phaeochromocytomas from patients with type I neurofibromatosis. Genes Chromosomes Cancer 1992;4(4):337-42. 225. Baysal BE, Ferrell RE, Willett-Brozick JE, Lawrence EC, Myssiorek D, Bosch A, van der Mey A, Taschner PE, Rubinstein WS, Myers EN, Richard CW, 3rd, Cornelisse CJ, Devilee P, Devlin B. Mutations in SDHD, a mitochondrial complex II gene, in hereditary paraganglioma. Science 2000;287(5454):848-51. 226. Astuti D, Douglas F, Lennard TW, Aligianis IA, Woodward ER, Evans DG, Eng C, Latif F, Maher ER. Germline SDHD mutation in familial phaeochromocytoma. Lancet 2001;357(9263):1181-2. 227. Astuti D, Latif F, Dallol A, Dahia PL, Douglas F, George E, Skold- berg F, Husebye ES, Eng C, Maher ER. Gene mutations in the succinate dehydrogenase subunit SDHB cause susceptibility to familial pheochromocytoma and to familial paraganglioma. Am J Hum Genet 2001;69(1):49-54. 228. Niemann S, Muller U. Mutations in SDHC cause autosomal dominant paraganglioma, type 3. Nat Genet 2000;26(3):268-70. 229. Hao HX, Khalimonchuk O, Schraders M, Dephoure N, Bayley JP, Kunst H, Devilee P, Cremers CW, Schiffman JD, Bentz BG, Gygi SP, Winge DR, Kremer H, Rutter J. SDH5, a gene required for flavination of succinate dehydrogenase, is mutated in paraganglioma. Science 2009;325(5944):1139-42. 230. Bayley JP, Kunst HP, Cascon A, Sampietro ML, Gaal J, Korpershoek E, Hinojar-Gutierrez A, Timmers HJ, Hoefsloot LH, Hermsen MA, Suarez C, Hussain AK, Vriends AH, Hes FJ, Jansen JC, Tops CM, Corssmit EP, de Knijff P, Lenders JW, Cremers CW, Devilee P, Din- jens WN, de Krijger RR, Robledo M. SDHAF2 mutations in familial and sporadic paraganglioma and phaeochromocytoma. The lancet oncology 2010;11(4):366-72. 231. Burnichon N, Briere JJ, Libe R, Vescovo L, Riviere J, Tissier F, Joua- nno E, Jeunemaitre X, Benit P, Tzagoloff A, Rustin P, Bertherat J, Favier J, Gimenez-Roqueplo AP. SDHA is a tumor suppressor gene causing paraganglioma. Hum Mol Genet 2010;19(15):3011-20.

96 232. van Nederveen FH, Korpershoek E, Lenders JW, de Krijger RR, Din- jens WN. Somatic SDHB mutation in an extraadrenal pheochromocytoma. The New England journal of medicine 2007;357(3):306-8. 233. Prowse AH, Webster AR, Richards FM, Richard S, Olschwang S, Resche F, Affara NA, Maher ER. Somatic inactivation of the VHL gene in Von Hippel-Lindau disease tumors. Am J Hum Genet 1997;60(4):765-71. 234. Akama H, Noshiro T, Kimura N, Shimizu K, Watanabe T, Shibukawa S, Nakai S, Miura W, Ito S, Miura Y. Multiple endocrine neoplasia type 2A with the identical somatic mutation in medullary thyroid carcinoma and pheochromocytoma without germline mutation at the corresponding site in the RET proto-oncogene. Internal medicine (Tokyo, Japan) 1999;38(2):145-9. 235. Gimm O, Armanios M, Dziema H, Neumann HP, Eng C. Somatic and occult germ-line mutations in SDHD, a mitochondrial complex II gene, in nonfamilial pheochromocytoma. Cancer Res 2000;60(24):6822-5. 236. Bender BU, Gutsche M, Glasker S, Muller B, Kirste G, Eng C, Neu- mann HP. Differential genetic alterations in von Hippel-Lindau syndrome-associated and sporadic pheochromocytomas. The Journal of clinical endocrinology and metabolism 2000;85(12):4568-74. 237. Hes FJ, Hoppener JW, Lips CJ. Clinical review 155: Pheochromocy- toma in Von Hippel-Lindau disease. The Journal of clinical endocrinology and metabolism 2003;88(3):969-74. 238. McWhinney SR, Boru G, Binkley PK, Peczkowska M, Januszewicz AA, Neumann HP, Eng C. Intronic single nucleotide polymorphisms in the RET protooncogene are associated with a subset of apparently sporadic pheochromocytoma and may modulate age of onset. The Journal of clinical endocrinology and metabolism 2003;88(10):4911- 6. 239. Dannenberg H, Komminoth P, Dinjens WN, Speel EJ, de Krijger RR. Molecular genetic alterations in adrenal and extra-adrenal pheochromocytomas and paragangliomas. Endocr Pathol 2003;14(4):329-50. 240. Gimenez-Roqueplo AP, Lehnert H, Mannelli M, Neumann H, Opoch- er G, Maher ER, Plouin PF. Phaeochromocytoma, new genes and screening strategies. Clinical endocrinology 2006;65(6):699-705. 241. Neumann HP, Pawlu C, Peczkowska M, Bausch B, McWhinney SR, Muresan M, Buchta M, Franke G, Klisch J, Bley TA, Hoegerle S, Boedeker CC, Opocher G, Schipper J, Januszewicz A, Eng C. Distinct clinical features of paraganglioma syndromes associated with SDHB and SDHD gene mutations. Jama 2004;292(8):943-51. 242. Lenders JW, Eisenhofer G, Mannelli M, Pacak K. Phaeochromocyto- ma. Lancet 2005;366(9486):665-75. 243. Benn DE, Robinson BG. Genetic basis of phaeochromocytoma and paraganglioma. Best practice & research 2006;20(3):435-50.

97 244. Martin TP, Irving RM, Maher ER. The genetics of paragangliomas: a review. Clin Otolaryngol 2007;32(1):7-11. 245. Dahia PL, Ross KN, Wright ME, Hayashida CY, Santagata S, Baron- tini M, Kung AL, Sanso G, Powers JF, Tischler AS, Hodin R, Heitrit- ter S, Moore F, Dluhy R, Sosa JA, Ocal IT, Benn DE, Marsh DJ, Ro- binson BG, Schneider K, Garber J, Arum SM, Korbonits M, Gross- man A, Pigny P, Toledo SP, Nose V, Li C, Stiles CD. A HIF1alpha regulatory loop links hypoxia and mitochondrial signals in pheochromocytomas. PLoS genetics 2005;1(1):72-80. 246. Carmeliet P, Dor Y, Herbert JM, Fukumura D, Brusselmans K, De- werchin M, Neeman M, Bono F, Abramovitch R, Maxwell P, Koch CJ, Ratcliffe P, Moons L, Jain RK, Collen D, Keshert E. Role of HIF- 1alpha in hypoxia-mediated apoptosis, cell proliferation and tumour angiogenesis. Nature 1998;394(6692):485-90. 247. Qin Y, Buddavarapu K, Dahia PL. Pheochromocytomas: From Genet- ic Diversity to New Paradigms. Horm Metab Res 2009. 248. Lee S, Nakamura E, Yang H, Wei W, Linggi MS, Sajan MP, Farese RV, Freeman RS, Carter BD, Kaelin WG, Jr., Schlisio S. Neuronal apoptosis linked to EglN3 prolyl hydroxylase and familial pheochromocytoma genes: Developmental culling and cancer. Cancer Cell 2005;8(2):155-67. 249. Ladroue C, Carcenac R, Leporrier M, Gad S, Le Hello C, Galateau- Salle F, Feunteun J, Pouyssegur J, Richard S, Gardie B. PHD2 mutation and congenital erythrocytosis with paraganglioma. The New Eng- land journal of medicine 2008;359(25):2685-92. 250. Qin Y, Yao L, King EE, Buddavarapu K, Lenci RE, Chocron ES, Lechleiter JD, Sass M, Aronin N, Schiavi F, Boaretto F, Opocher G, Toledo RA, Toledo SP, Stiles C, Aguiar RC, Dahia PL. Germline mutations in TMEM127 confer susceptibility to pheochromocytoma. Nat Genet 2010;42(3):229-33. 251. Jarbo C, Buckley PG, Piotrowski A, Mantripragada KK, Benetkiewicz M, de Stahl TD, Langford CF, Gregory SG, Dralle H, Gimm O, Back- dahl M, Geli J, Larsson C, Westin G, Akerstrom G, Dumanski JP. De- tailed assessment of chromosome 22 aberrations in sporadic pheochromocytoma using array-CGH. Int J Cancer 2005. 252. Aarts, Aarts M, Dannenberg H, deLeeuw RJ, van Nederveen FH, Verhofstad AA, Lenders JW, Dinjens WN, Speel EJ, Lam WL, de Krijger RR. Microarray-based CGH of sporadic and syndrome-related pheochromocytomas using a 0.1-0.2 Mb bacterial artificial chromosome array spanning chromosome arm 1p. Genes Chromosomes Can- cer 2006;45(1):83-93. 253. van Nederveen F, Korpershoek E, Deleeuw R, Verhofstad AA, Lend- ers JW, Dinjens W, Lam W, de Krijger R. Array-CGH in sporadic benign pheochromocytomas. Endocr Relat Cancer 2009(16):505-13.

98 254. Petri BJ, Speel EJ, Korpershoek E, Claessen SM, van Nederveen FH, Giesen V, Dannenberg H, van der Harst E, Dinjens WN, de Krijger RR. Frequent loss of 17p, but no p53 mutations or protein overexpression in benign and malignant pheochromocytomas. Mod Pathol 2008;21(4):407-13. 255. Tonks ID, Mould AW, Schroder WA, Cotterill A, Hayward NK, Walker GJ, Kay GF. Dual loss of rb1 and Trp53 in the adrenal medulla leads to spontaneous pheochromocytoma. Neoplasia (New York, NY 2010;12(3):235-43. 256. Eisenhofer G, Huynh TT, Pacak K, Brouwers FM, Walther MM, Li- nehan WM, Munson PJ, Mannelli M, Goldstein DS, Elkahloun AG. Distinct gene expression profiles in norepinephrine- and epinephrine- producing hereditary and sporadic pheochromocytomas: activation of hypoxia-driven angiogenic pathways in von Hippel-Lindau syndrome. Endocr Relat Cancer 2004;11(4):897-911. 257. Fassnacht M, Weismann D, Ebert S, Adam P, Zink M, Beuschlein F, Hahner S, Allolio B. AKT is highly phosphorylated in pheochromocytomas but not in benign adrenocortical tumors. The Journal of clinical endocrinology and metabolism 2005;90(7):4366-70. 258. Fottner C, Minnemann T, Kalmbach S, Weber MM. Overexpression of the insulin-like growth factor I receptor in human pheochromocytomas. Journal of molecular endocrinology 2006;36(2):279-87. 259. Lai EW, Joshi BH, Martiniova L, Dogra R, Fujisawa T, Leland P, de Krijger RR, Lubensky IA, Elkahloun AG, Morris JC, Puri RK, Pacak K. Overexpression of interleukin-13 receptor-alpha2 in neuroendocrine malignant pheochromocytoma: a novel target for receptor di- rected anti-cancer therapy. The Journal of clinical endocrinology and metabolism 2009;94(8):2952-7. 260. Muscarella P, Bloomston M, Brewer AR, Mahajan A, Frankel WL, Ellison EC, Farrar WB, Weghorst CM, Li J. Expression of the p16INK4A/Cdkn2a gene is prevalently downregulated in human pheochromocytoma tumor specimens. Gene expression 2008;14(4):207-16. 261. Kiss NB, Geli J, Lundberg F, Avci C, Velazquez-Fernandez D, Ha- shemi J, Weber G, Hoog A, Ekstrom TJ, Backdahl M, Larsson C. Me- thylation of the p16INK4A promoter is associated with malignant behavior in abdominal extra-adrenal paragangliomas but not pheochromocytomas. Endocr Relat Cancer 2008;15(2):609-21. 262. mGeli J, Nord B, Frisk T, Edstrom Elder E, Ekstrom TJ, Carling T, Backdahl M, Larsson C. Deletions and altered expression of the RIZ1 tumour suppressor gene in 1p36 in pheochromocytomas and abdominal paragangliomas. Int J Oncol 2005;26(5):1385-91. 263. Margetts CD, Astuti D, Gentle DC, Cooper WN, Cascon A, Cat- chpoole D, Robledo M, Neumann HP, Latif F, Maher ER. Epigenetic analysis of HIC1, CASP8, FLIP, TSP1, DCR1, DCR2, DR4, DR5,

99 KvDMR1, H19 and preferential 11p15.5 maternal-allele loss in von Hippel-Lindau and sporadic phaeochromocytomas. Endocr Relat Cancer 2005;12(1):161-72. 264. Dammann R, Schagdarsurengin U, Seidel C, Trumpler C, Hoang-Vu C, Gimm O, Dralle H, Pfeifer GP, Brauckhoff M. Frequent promoter methylation of tumor-related genes in sporadic and men2-associated pheochromocytomas. Exp Clin Endocrinol Diabetes 2005;113(1):1-7. 265. Astuti D, Agathanggelou A, Honorio S, Dallol A, Martinsson T, Kogner P, Cummins C, Neumann HP, Voutilainen R, Dahia P, Eng C, Maher ER, Latif F. RASSF1A promoter region CpG island hypermethylation in phaeochromocytomas and neuroblastoma tumours. Onco- gene 2001;20(51):7573-7. 266. Geli J, Kiss N, Lanner F, Foukakis T, Natalishvili N, Larsson O, Kogner P, Hoog A, Clark GJ, Ekstrom TJ, Backdahl M, Farnebo F, Larsson C. The Ras effectors NORE1A and RASSF1A are frequently inactivated in pheochromocytoma and abdominal paraganglioma. En- docr Relat Cancer 2007;14(1):125-34. 267. Margetts CD, Morris M, Astuti D, Gentle DC, Cascon A, McRonald FE, Catchpoole D, Robledo M, Neumann HP, Latif F, Maher ER. Evaluation of a functional epigenetic approach to identify promoter region methylation in phaeochromocytoma and neuroblastoma. En- docr Relat Cancer 2008;15(3):777-86. 268. Kallioniemi A, Kallioniemi OP, Sudar D, Rutovitz D, Gray JW, Waldman F, Pinkel D. Comparative genomic hybridization for molecular cytogenetic analysis of solid tumors. Science 1992;258(5083):818-21. 269. Pinkel D, Segraves R, Sudar D, Clark S, Poole I, Kowbel D, Collins C, Kuo WL, Chen C, Zhai Y, Dairkee SH, Ljung BM, Gray JW, Al- bertson DG. High resolution analysis of DNA copy number variation using comparative genomic hybridization to microarrays. Nat Genet 1998;20(2):207-11. 270. Solinas-Toldo S, Lampel S, Stilgenbauer S, Nickolenko J, Benner A, Dohner H, Cremer T, Lichter P. Matrix-based comparative genomic hybridization: biochips to screen for genomic imbalances. Genes Chromosomes Cancer 1997;20(4):399-407. 271. Emanuel BS, Saitta SC. From microscopes to microarrays: dissecting recurrent chromosomal rearrangements. Nat Rev Genet 2007;8(11):869-83. 272. Ishkanian AS, Malloff CA, Watson SK, DeLeeuw RJ, Chi B, Coe BP, Snijders A, Albertson DG, Pinkel D, Marra MA, Ling V, MacAulay C, Lam WL. A tiling resolution DNA microarray with complete coverage of the human genome. Nat Genet 2004;36(3):299-303. 273. Andersson R, Bruder CE, Piotrowski A, Menzel U, Nord H, Sandgren J, Hvidsten TR, Diaz de Stahl T, Dumanski JP, Komorowski J. A

100 segmental maximum a posteriori approach to genome-wide copy number profiling. Bioinformatics 2008;24(6):751-8. 274. Solomon MJ, Larsen PL, Varshavsky A. Mapping protein-DNA interactions in vivo with formaldehyde: evidence that histone H4 is retained on a highly transcribed gene. Cell 1988;53(6):937-47. 275. Pillai S, Chellappan SP. ChIP on chip assays: genome-wide analysis of transcription factor binding and histone modifications. Methods in molecular biology (Clifton, NJ 2009;523:341-66. 276. Pillai S, Dasgupta P, Chellappan SP. Chromatin immunoprecipitation assays: analyzing transcription factor binding and histone modifications in vivo. Methods in molecular biology (Clifton, NJ 2009;523:323-39. 277. Wales MM, Biel MA, el Deiry W, Nelkin BD, Issa JP, Cavenee WK, Kuerbitz SJ, Baylin SB. p53 activates expression of HIC-1, a new candidate tumour suppressor gene on 17p13.3. Nat Med 1995;1(6):570-7. 278. Fleuriel C, Touka M, Boulay G, Guerardel C, Rood BR, Leprince D. HIC1 (Hypermethylated in Cancer 1) epigenetic silencing in tumors. The international journal of biochemistry & cell biology 2009;41(1):26-33. 279. Bagchi A, Mills AA. The quest for the 1p36 tumor suppressor. Cancer Res 2008;68(8):2551-6. 280. Mulero-Navarro S, Esteller M. Chromatin remodeling factor CHD5 is silenced by promoter CpG island hypermethylation in human cancer. Epigenetics 2008;3(4):210-5. 281. Gorringe KL, Choong DY, Williams LH, Ramakrishna M, Sridhar A, Qiu W, Bearfoot JL, Campbell IG. Mutation and methylation analysis of the chromodomain-helicase-DNA binding 5 gene in ovarian cancer. Neoplasia (New York, NY 2008;10(11):1253-8. 282. Cooper DN, Krawczak M, Antonorakis SE. The nature and mechanisms of human gene mutation. 7 ed. New York: McGraw-Hill; 1995. 283. Sun F, Chan E, Wu Z, Yang X, Marquez VE, Yu Q. Combinatorial pharmacologic approaches target EZH2-mediated gene repression in breast cancer cells. Molecular cancer therapeutics 2009;8(12):3191- 202. 284. Ellis L, Atadja PW, Johnstone RW. Epigenetics in cancer: targeting chromatin modifications. Molecular cancer therapeutics 2009;8(6):1409-20. 285. Tan J, Yang X, Zhuang L, Jiang X, Chen W, Lee PL, Karuturi RK, Tan PB, Liu ET, Yu Q. Pharmacologic disruption of Polycomb- repressive complex 2-mediated gene repression selectively induces apoptosis in cancer cells. Genes & development 2007;21(9):1050-63. 286. Meyer-Rochow GY, Jackson NE, Conaglen JV, Whittle DE, Kunni- malaiyaan M, Chen H, Westin G, Sandgren J, Stalberg P, Khanafshar E, Shibru D, Duh QY, Clark OH, Kebebew E, Gill A, Clifton-Bligh R,

101 Robinson BG, Benn DE, Sidhu SB. MicroRNA profiling of benign and malignant pheochromocytoma identifies novel diagnostic and therapeutic targets. Endocr Relat Cancer 2010. 287. Wooster R, Bachman KE. Catalogue, cause, complexity and cure; the many uses of cancer genome sequence. Current opinion in genetics & development 2010;20(3):336-41.

102