Epigenomics and cancer

Abstract

Cancer genes are recognized by their altered gene expression or activity, or both, leading to an abnormal phenotype. Nearly every tumor type presents an enormous complexity of altered gene functions, including activation of growth-promoting genes as well as silencing of genes with tumor growthsuppressing functions, all contributing to uncontrolled growth. These changes provide the cell with a competitive growth advantage that is realized through at least five cancercell phenotypes: enhanced cell division, resistance to apoptosis, sustained angiogenesis, invasion of tissues and metastasis, and evasion of antitumor immune responses (reviewed in Reference 1). Traditionally, only mutated genes have been considered as candidate cancer genes. However, clearly many more genes present altered gene expression in cancer cells than are mutated [2]. Epigenetic changes, mainly DNA methylation and, more recently, modification of histones, are now recognized as additional mechanisms with a major contribution to the malignant phenotype. Epigenetic inheritance involves the transmission of information not encoded in DNA sequences from cell to daughter cell or from generation to generation. Covalent modifications of the DNA or its packaging histones are responsible for transmitting epigenetic information. Functionally, epigenetic marks on the DNA and histones act to regulate gene expression, silence the activity of transposable elements and stabilize adjustments of gene dosage, as seen in X-chromosome inactivation and genomic imprinting. In mammals, epigenetic regulation is crucial for a variety of different processes such as development, cell differentiation, and proliferation. Additionally, epigenetic state of the genome is modulated by factors such as disease, nutrition, age, and sex. Whereas genetic alterations leave a permanent print in the genome, epigenetic alterations are reversible and the enzymes responsible for their maintenance are the potential target for a number of therapeutic compounds. Epigenetic modifications constitute the basis to establish the profiles of gene expression and nuclear organization for a given set of genomic information. This information determines cell type identity. Basically, cells encode their epigenetic information in two groups of molecules: DNA and histones. In DNA, methylation of the 5-position of cytosine in CpG dinucleotides is the most common epigenetic modification [3]. The methyl group is transferred from S-adenosy lmethionine to the C-5 position of cytosine by a family of DNA methyltransferases (DNMT). DNA methylation occurs almost exclusively at adjacent cytosine and guanine nucleotides in the DNA (CpG) nucleotides. CpG are unevenly distributed throughout the vertebrate genome, where this dinucleotide is relatively uncommon and has a tendency to cluster in regions known as CpG islands [4], many of which are coincident with the promoter of protein-coding genes. Most dispersed CpG in the genome are methylated, unlike in CpG islands, where methylation occurs rarely in normal cells [4] and results in transcriptional repression [5]. This situation is restricted to a small number of genes, including imprinted genes, X-chromosome genes in women, and a few tissuespecific genes whose expression is only required for a short period. The first observation of DNA methylation aberrations in human cancer was the finding that tumors were globally hypomethylated [6]. This discovery was found only 1 year after the first oncogene mutation was discovered in the Hras in a human primary tumor. The idea that the genome of the cancer cell undergoes a reduction of its 5-methylcytosine content compared with the normal tissue has been firmly corroborated [7, 8]; however, it does not associate with overexpression of oncogenes as originally thought and may be related with the generation of chromosomal instability. Not only is global DNA hypomethylation a common hallmark in cancer but also, paradoxically, hypermethylation of promoter regions of tumor suppressor genes (TSG) is also. To the best of our knowledge, the first discovery of methylation in a CpG island of a TSG in a human cancer was that of the retinoblastoma (Rb) gene in 1989 [9]. Not until 1994, the idea that CpG island promoter hypermethylation could be a mechanism to inactivate genes in cancer was fully restored as a result of the discovery that the Von Hippel-Lindau (VHL) gene also undergoes methylation-associated inactivation [10]. The origin of the current period of research in cancer epigenetic silencing was perhaps the discovery that CpG island hypermethylation was a common mechanism of inactivation of the TSG p16INK4a in human cancer [11, 12]. For many of these hypermethylated TSGs, it has been shown that their re-expression in tumor cells by demethylating drugs can lead to suppression of cell growth or altered sensitivity to existing anticancer therapies. As compounds have been identified that can readily reverse epigenetic silencing, there is increasing interest in epigenetic regulation of gene expression as a basis for new approaches to cancer treatment. For many years, epigenetic research focused on DNA methylation; this is changing and considerable attention is being given to histone modifications as well. DNA is wrapped around an octamer of histones called a nucleosome that constitutes the building unit of chromatin. Histone tails receive epigenetic information through a complex set of posttranslational modifications [13] including methylation, acetylation, phosphorylation, and sumoylation. Evidence exists that histone modifications help to determine higher-order chromatin structures, which may in turn influence the transcriptional status (e.g., see References 14 and 15). Additionally, there is increasing evidence that characteristic modifications patterns (the "histone code") on histone tails are involved in gene regulation through changes in chromatin structure and condensation. The histone code is recognized by effector proteins that bind to the nucleosomes inducing changes on gene expression [16]. For example, methylation of H3 at lysine-4 [17] or arginine- 17 [18] is closely linked to transcriptional competence, whereas methylation of H3 at lysine-9 or H4 at lysine-20 is associated with transcriptional repression [19, 20]. Initially, aberrations in post-translational modifications of histones in cancer cells were only shown to occur at individual promoters. These changes were reported to be associated with the presence of methyl-binding domain (MBD) proteins [21, 22]. In this context, hypermethylation of the promoter CpG islands of TSG was thought to be mechanistically linked to gene silencing through the recruitment of MBDs. The binding of MBDs to hypermethylated promoters would be followed by a change in the pattern of histone modifications that, in turn, would lead to a change in the chromatin structure compatible with gene inactivation. Furthermore, we have recently characterized post-translational modifications of histone H4 at a global level in a comprehensive panel of normal tissues, cancer cell lines, and primary tumors [23] and found that the global loss of monoacetylation and trimethylation of histone H4 is a common hallmark of human tumor cells. Besides having a direct effect on transcriptional activity, DNA methylation and histone modifications also play a key role in organizing nuclear architecture [15, 24], which in turn is also involved in regulating transcription and other nuclear processes. Therefore, epigenetic modifications are essential for defining the cellular transcriptome at several levels. Aberrant changes in the pattern of epigenetic modifications result in altered nuclear activity, and thereby altered transcriptome, transforming the identity of the cell. Although the importance of altered epigenetic regulation in tumorigenesis is clearly proven, little is known about its extent and genomic distribution. Epigenomics search to define the epigenetic pattern in a genome-wide scale. This term encompasses not only whole-genome studies of epigenetic processes but also the identification of the characteristic DNA sequences that specify where the epigenetic processes are targeted. Because it was realized that CpG dinucleotides in mammals represent the target for the covalent modification of DNA, it has been apparent that DNA sequence characteristic can influence that targeting of epigenetic processes. Historically, technology has limited large-scale approaches to epigenomics, but the emergence of highly reproducible quantitative highthroughput microarray technology is allowing nearly all epigenomics research to be read on microarray platforms. The field is nascent at present, and efforts to develop it include arraybased methylation analysis, array-based hybridization using probes prepared by immunoprecipitation with antibodies (Ab) to modified histones (so-called ChIP-on-CHIP), and highthroughput allele-specific expression analysis (for excellent reviews on epigenomics see References 25 and 26). In the following sections, we analyze all these aspects in greater detail. © 2008 Humana Press Inc.