[Frontiers in Bioscience 16, 2682-2694, June 1, 2011]

Epigenetic regulation in cancer development

Elisa Caffarelli1, Patrizia Filetici1

1Istituto di Biologia e Patologia Molecolari, Laboratorio Acidi Nucleici, Sapienza Universita di Roma, P.le A.Moro 5, 00185 Roma-Italy

TABLE OF CONTENTS

1. Abstract
2. Introduction
2.1. A comprehensive definition of Epigenetic
3. Epigenetic and carcinogenesis
3.1. DNA Methylation
3.2. Chromatin signalling
3.3. Epigenetic effectors and histone PTMs
4. Altered epigenetic patterns in cancer
4.1. Global histone PTMs in cancer tissues
5. miRNAs take their place in the epigenetic world
6. Dysregulation of miRNA expression associated to carcinogenesis
6.1. Tumor suppressor and/or oncogenic miRNAs
6.2. miRNAs as tumor diagnostic tools
7. Epigenetic control of miRNA genes
8. miRNA control of epigenetic mechanisms
9. Summary and perspective
10. Acknowledgements
11. References

1. ABSTRACT

From an operational definition of epigenetic, we move to provide the reader a general but comprehensive description of epigenetic phenomena that often lead to cell transformation. The last decade has, in fact, seen novel players involved in the regulation of gene expression. Not only protein factors but also a number of chromatin modifiers and remodelling proteins, which regulate the level of compaction of the genome through a variety of post-translational modifications deposed on histone tails or on DNA itself. Meanwhile, the discovery of tiny RNAs, of only 21-23 nucleotides in length, has brought to the attention their role as key regulators in the cell, being able to direct differentiation programs and function as oncogenes or oncosuppressors. In this general compendium, we aim to describe main cellular functions that through an epigenetic or epigenetic associated mechanism have been found to be directly implicated in cancerogenesis.

2. INTRODUCTION

2.1. A comprehensive definition of Epigenetic

"An epigenetic trait is a stably inherited phenotype resulting from changes in a chromosome without alteration in the DNA sequence". This operational definition, proposed in a world conference (1), takes into account the heritability of a phenotype, which does not rely on a genetic modification of DNA sequence. Accordingly, it was proposed that three different classes of signals, operating in a concerted manner, create heritable epigenetic state. The first signal, called Epigenator, arises from the environment and acts triggering intracellular pathways; the second signal, the Epigenetic Initiator, reacts to the Epigenator determining the precise location of the epigenetic chromatin environment and, finally, the Epigenetic Maintainer, as an internal signal, sustains the chromatin state through subsequent generations.

The Initiator, which responds to transient external stimuli defining the region of chromosome where the chromatin state should be established, may be a DNA binding factor, a noncoding RNA or other effectors able to determine a precise epigenetic chromatin state. In the cell, the Initiator may persist and propagate its function in combination with the Maintainer, which holds a defined epigenetic state up. Maintainer involves different pathways like DNA methylation, histone modifications (PTMs), deposition of histone variants and nucleosome localization. All these features, that require Initiator and are DNA sequence-independent, need to be exactly positioned in the genome. Deregulation of such processes may lead to genetic syndromes, cellular aging or cancer transformation.

Still, definition of epigenetic may be broader than this, and not necessarily linked to hereditability. The US National Institutes of Health (2009) states that "Epigenetics refers to both heritable changes in gene activity and expression (in the progeny of cells or of individuals) and also stable, long-term, alterations in the transcriptional potential of a cell that are not necessarily heritable". Overall, we can assess that epigenetic processes lead to a stable alteration of gene expression. Collectively, they include several events like cytosine methylation, post-translational modifications of histone and non-histone proteins, chromatin remodeling, nucleosome positioning and RNA-based mechanisms (2).

3. EPIGENETIC AND CARCINOGENESIS

In the last years, a crucial issue is the comprehension of the molecular mechanisms underlying tumorigenesis. This is a multi-faceted process, mainly based on the dys-regulation of two antagonistic classes of genes: tumor suppressor genes, which inhibit cell growth and survival, and oncogenes, that promote cell proliferation and mitosis through cell signalling pathways. It is now clear that activation and repression of cancer-related genes can be due to both genetic and epigenetic alterations. The genetic changes consist of loss or amplification of cancer-associated genes, whereas the epigenetic modifications may involve various mechanisms and imply different pathways. Several studies reveal that epigenetic modifications may result in activation of oncogenes or silencing of tumor suppressors, which end with cell transformation and uncontrolled proliferation. Often, in cancer cells several mutated or translocated chromatin modifiers together with aberrant DNA methylation of single genes or genomic regions have been found. Histone marks and modifications are also often altered, and a relevant question is whether specific PTMs and chromatin states are peculiar of disease and oncogenic traits. Therefore, advances in identification of recurrent epigenetic traits occurring in human cancer will help in set up predictive tools for cancer prognosis and future disease outcome.

Recent studies indicated that also RNA-mediated mechanisms are involved in more widespread epigenetic regulation. In particular, several epigenetic mechanisms include regulation of gene expression mediated by long non coding RNAs, some of which interact with chromatin modifying complexes (3, 4, 5), and by small non coding RNAs, as microRNAs (miRNAs) that regulate components of the epigenetic machinery. Notably, dysregulation of such RNA-regulators may have profound impact on the establishment and maintenance of the epigenetic landscapes characteristic of differentiated cells.

3.1. DNA Methylation

DNA methylation is the most widely studied epigenetic modification in mammals; it is a trait that is heritably propagated and consists of modification, namely, methylation of cytosine on DNA at CpG target sites (6). DNA methylation influences several processes like gene regulation and nucleosome positioning, imprinting and X chromosome inactivation (7). Methylation of CpG can act directly, by preventing the binding of transcription factors, or indirectly, by recruitment of methyl-binding chromodomain proteins (8) and of silencing structural proteins, like HP1, thus reinforcing a repressive function on vaste genomic regions (9). Such process stabilises a repressive heterochromatinization, which, may further silence gene expression by recruitment of HDACs (10), that induce a generalized chromatin deacetylation.

Often, DNA methylation may occur not only at CpG, but also at repetitive DNA sequences like centromeres, transposons and telomeres. In these cases, such DNA modification leads to genome instability causing chromosome rearrangements and translocations. Notably, alteration of DNA methylation induces the aberrant activation or repression of various signaling pathways triggering cell transformation. The "two hits" model (11) proposes that, in cancer, the genetic mutation of one allele is reinforced by methylation of the second resulting in loss of heterozygosity. Since epimutations are inherited, they can survive and selectively expand in a rapidly growing cell population, thus conferring a selective advantage over the normal propagating cells. This epigenetic event may occur very early in cancer development and represents the first hit for uncontrolled cancer cell expansion (12).

CpG dinucleotides are clustered in short CpG islands, often localized at 5' end of genes, and are associated to almost 60% of human promoters (13). While some CpG are methylated, the vaste majority remain unmethylated; however, the dynamic equilibrium between methylated and unmethylated DNA states is crucial for development and differentiation programs (14).

DNA methylation works in concert with silencing processes acting through repressive histone marks and produces a higher ordered compaction of chromatin. In addition, histone methyl marks may provide a signal for further recruitment of chromatin associated repressors. Histone H3-K9me3, is, for example, the target for the recruitment of HP1 or MECP2 (15), whereas, histone H3-K27me, interacts with Polycomb Repressive Complexes (PRCs) and recruits HDAC to silence gene expression. Hypomehtylation leads to the activation of growth-promoting oncogenes such as R-Ras and MASPIN in gastric cancer, S-100 in colon cancer and MAGE in melanoma (16). In Wilms' tumor, loss of heterozygosity is associated to an early event of DNA hypomethylation inducing a further loss of imprinting of IGF2 gene (17). The hypermethylation also contributes to tumorigenesis by repressing suppressors. Several genes have been associated to heavy CpG island hypermethylation; from the first report on Retinoblastoma, Rb (18), various other tumor suppressor genes, including p16, MLH1 and BRCA1, have been shown to undergo tumor-specific silencing by hypermethylation (19). DNA methylation can also exert its function through repression of transcription factor genes involved in the activation of oncosuppressors, like RUNX3 in esophageal cancer (20) and GATA-4 and GATA-5 in colorectal and gastric cancers (21). Oncogenic transcription factors, like the aberrant PML-RAR fusion protein found in acute promyelocytic leukemia (22) may recruit DNA Methyl-Transferases (DNMTs) to specific genes where they induce a tumor specific CpG island methylation. The presence of methylation on a subset of CpG islands corresponding to known polycomb targets involved in differentiation, was found in colorectal cancer providing a direct link between the process of DNA methylation and cellulare differentiation (23).

3.2. Chromatin signalling

Chromatin is organized in nucleosomes, nucleoprotein complexes, which are the fundamental units that contribute to the structural dynamic of the genome. Distinct chromatin structural states, affecting the accessibility of DNA to protein complexes and transcriptional machinery, regulate gene expression. Chromatin is, therefore, fluctuating in a dynamic equilibrium between heterochromatin, the compacted, silenced and transcriptionally inactive state, and the euchromatin, the decondensed and active state bearing spaced nucleosomes. In the nucleosome, the DNA is wrapped around the fundamental repeat of chromatin, represented by the histone octamer. The canonical octamer is formed by histones (H3, H4, H2A and H2B)2 and consists of a central globular region from which protrude prolonged external N- termini; these represent the main place of deposition of a variety of post-translational modifications (PTMs) (24), like acetylation and methylation of lysines, methylation of arginines and phosphorylation of serines. Histone tails are, therefore, signal platforms where key epigenetic marks are deposed producing subsequent structural modulation and variation of the chromatin compaction. Overall, PTMs (25) work through in cis activity, which mainly affects processes like the chromatin remodelling, nucleosome positioning and regulation of gene transcription, or in trans activity, dealing with the interaction of chromatin associated proteins. Such proteins interact at specific regions of the genome with PTMs through chromatin reading modules that recognize single modified residues on histone tails (26).

3.3. Epigenetic effectors and histone PTMs

The epigenetic effectors or modifiers are enzymes, DNMTs, HMTs, HATs, HDACs, often member of multi-protein complexes, whose catalytic activity is responsible for post-translational chemical modifications. DNMT family includes three different enzymes: DNMT1 (maintenance DNMT), which preserves the methylation pattern throughout each cell division, and DNMT-3a and 3b (de novo DNMTs), which transfer a methyl group to previously unmethylated genomic regions (27). Acetyltransferases (HATs) are divided into three main families, GNAT, MYST, and CBP/p300. In general these enzymes modify more than one lysine but some limited specificity can be detected for some enzymes. There are three distinct families of histone deacetylases (HDACs): the class I and class II histone deacetylases and the class III NAD-dependant enzymes of the Sir family. They are involved in multiple signaling pathways and are present in numerous repressive chromatin complexes. Lysine methyltransferases (HMTs) have enormous specificity compared to acetyltransferases. They usually modify one single lysine on a single histone and their output can be either activation or repression of transcription. Like lysine methylation, arginine methylation can be either activatory or repressive for transcription, and the enzymes protein arginine methyltransferases, (PRMT's) are recruited to promoters by transcription factors. PTMs are able to change the functional properties of the chromatin fiber, which, in turn, affects cellular processes leading to activation or silencing of gene expression, recombination, repair and replication. A histone code of PTMs (28), deposed at specific sites on the same (intra) or different (inter) histones tails, is finely regulated and is often cross regulated by reciprocal activity of PTMs, like acetylation, phosphorylation, methylation ubiquitination, sumoylation, proline isomerization and ADP-ribosylation (29). As an example of an inter-nucleosomal cross-talk is the requirement of histone H2B-K123 monoubiquitination, for methylation of histone H3-K4 and H3-K79 during transcription. Also histone H3 acetylation was shown to stimulate the other positive mark H3-K4 methylation. From these and other data it has been, therefore, proposed a histone code hypothesis where activating marks may reciprocally regulate the deposition of other modifications on the same histone tail or across tails of different histones (Figure 1).

Among histone PTMs, acetylation is a bona fide positive mark; it neutralizes the positive charges of lysine side chains, thus reducing the strength of the binding of histone tails to DNA. This mechanism leads to the opening of chromatin structure, enabling nucleosome eviction and/or repositioning. Methylation of lysines like histone H3-K4, K36 (30), K79 and H4-K20, and arginines H3-R2, R17, R26 and H4-R3, represent activating marks, whereas methylation of H3-K9, K27 and H4-K20 are associated to repression and may be classified as repressive marks (31, 32).

In gene silencing, the nucleosomes must be deacetylated to prevent cryptic initiation of transcription. It was reported that methylation of histone H3-K36me provides the signal for the recruitment of the HDAC Rpd3 complex (33), which through deacetylation compacts chromatin and represses gene expression. In sum, histone PTMs regulate the expression of disease genes by modifying individual promoters like that of the cell cycle dependent cyclin E gene, whose downregulation is linked to ipoacetylation though aberrant recruitment of HDAC1 by retinoblastoma (Rb) (34).

4. ALTERED EPIGENETIC PATTERNS IN CANCER

Cancer may evolve from combinatorial convergence of genetic and epigenetic abnormalities (35). Altered epigenetic landscape is characterized by aberrant CpG island methylation and deacetylation and/or methylation of histones (36, 37). In case of mutation of a single allele, the silencing of the second one may induce the hit, resulting in loss of heterozygosity and inactivation of tumor suppressor genes. In addition, silencing of oncosuppressors may cooperate with oncogenic mutations and induce tumor development. Accordingly, loss of heterozygosity has been recently considered as a relevant marker for diagnosis and prognosis of cancer. The epigenetic aberration linked to chromosome instability may also contribute to the loss of heterozygosity increasing the frequency of aneuploidy, which is an hallmark of many cancer types.

In several tumours, dysregulation of epigenetic modifiers produce alteration in PTMs pattern on critical promoters or at bulk genomic level with consequent uncontrolled proliferation of undifferentiated totipotent cell sub-populations. Immunostaining of cancer nuclei showed great heterogeneity in bulk levels of histone modifications in tissue specimens (38); accordingly, the identification of aberrant epigenetic states of chromating may represent novel predictive tools for human cancer.

In leukaemia, translocations involving Mixed Linear Leukaemia (MLL) gene leads to the loss of H3K4 methyltransferase activity, which contributes to transform haematopoietic into leukaemia stem cells with poor prognosis (39). Similarly, negative regulators like Polycomb Group (PcG) complexes are engaged in the targeted repression of genes during cell cycle. Two different Polycomb complexes, PRC1 and PRC2 have been shown, in fact, to be involved in different cancers. Up-regulation of EZH2, the PRC2-H3K27 methyltransferase, is found in mantle cell lymphoma (40), breast (41) and prostate cancers (42). In addition, PRC complex can be aberrantly recruited, by PML-RAR alpha oncofusion protein, at tumour suppressor genes thus inducing their repression in leukemia (43). Another component of PRC1, RING1, which monoubiquitylates histone H2A-K119, is upregulated in prostate cancer (44).

Acetylation counteracted by deacetylation is achieved by a fine equilibrium between opposing HAT and HDAC catalytic activity crucial for cell regulation. Deregulation of HATs activity is often found in cancer; this is the case of the chimeric fusion of two HAT catalytic domains MOZ-p300 involved in leukemogenesis in Acute Monocytic Leukemia (AML) (45). Collectively, these and other reports demonstrate that untargeted aberrant acetylation and unbalanced deacetylation are recurrently found, and become negative regulators of tumor suppressor genes (46). P ALIGN="JUSTIFY">4.1. Analysis of global histone PTMs in cancer tissues

Aberration of PTMs generally occur at single gene promoters where they regulate transcription and affect the binding of transcription factors. Nonetheless, a global genomic assay of bulk post-translational modifications of histones might represent an innovative approach for screening normal and cancer tissues. In the clinical practise it is, in fact, not easy to evaluate and classify cancer cells. Several cell types often elude the molecular analysis or lack of a clear classification method, as in the case of prostate cancer which is difficult to classify but still represents the second cause of death in US.

It was reported that the pattern of histone modification was indeed a prognostic feature in some tumors like prostate, kidney, lung, gastric ovarian and breast cancer (Figure 1). Locus specific changes in histone acetylation or methylation was correlated to the expression of critical genes in highly aggressive pancreatic adenocarcinoma (47).

In human breast carcinomas, global histone modification profile represented a tool for the identification of cancer phenotype and a prognostic sign of patient outcome (48). It was reported that H4K16ac was very low or absent in the majority of breast cancer cases. High relative levels of global histone acetylation and methylation were associated with a favorable prognosis. On the other hand, low levels of lysine acetylation (H3-K9ac, K18ac, and H4-K12ac), lysine methylation (H3-K4me2 and H4-K20me3), and arginine methylation (H4-R3me2) were found in carcinomas of poorer prognostic subtypes, including basal carcinomas and HER-2-positive tumors. Other reports demonstrated that also the level of acetylated H3-K9, K18 and H4-K12 and dimethylated H4-R3 and H3-K4 in prostate cancers were prognostic signatures (38). Loss of histone H4-K16ac and K20me3 were clearly associated to DNA hypomethylation at non coding repetitive sequences (49). In prostate cancer, lower to higher risk subtypes were compared and grouped on the basis of a statistical analysis of histone PTMs revealed by immunostaining (38). Although this type of analysis must be further extended, we may assume that this approach might be useful not only for cancer classification but also for predicting response to a specific drug-treatment.

5. miRNAs TAKE THEIR PLACE IN THE EPIGENETIC WORLD

For their mode of action, miRNAs provide an important layer of epigenetic information according to the broader definition of epigenetics (2): they modify gene expression without modifying DNA sequences.

The effectiveness of these small non-coding RNAs (approximately 21-nucleotide-long) in silencing gene expression at the post-transcriptional level in eukaryotic cells resides in their ability to act through very simple but highly specific mechanisms. miRNAs recognize their mRNA targets by sequence-specific base-pairing and, depending on the degree of complementarity, they can induce mRNA degradation (perfect pairing) or translational repression (imperfect pairing). Through this procedure, they function as molecular guides that address the multi-protein effector complex (miRNP) on the specific mRNA targets, preventing gene expression. Recent studies using high-throughput proteomics approaches (50; 51), highlighted the ability of individual miRNAs to affect hundreds of proteins in humans. This pleiotropic action allow them to simultaneously tune up the activity of multiple genes, which in turn often control a biologically coordinated genetic program. For this feature, miRNAs may globally affect entire pathways exerting a significant impact on cell fate.

On the other hand, it has been observed that a single mRNA can be controlled by multiple miRNAs, which can act combinatorially. Therefore, even if individual miRNAs suppress their targets only moderately, they can exert broad and strong effect on gene expression.Recently, more than five hundred miRNAs have been identified in humans, but at least a thousand have been predicted by bioinformatic analyses (52; 53). Functional studies indicate that they are crucial players in vital processes such as development, differentiation, cell proliferation and cell death, regulating the activity of approximately 30% of all protein-coding genes in mammals. Notably, dys-regulation of miRNA expression has been associated to developmental defects and to the onset and/or progression of carcinogenesis in humans. Accordingly, miRNAs are aberrantly expressed in a variety of cancers (54)

6. DYSREGULATION OF miRNA EXPRESSION IS ASSOCIATED TO CARCINOGENESIS

The first correlation between miRNA aberrant expression and cancer derived from the observation that miR-15a and miR-16-1, which are down-regulated in the cancer samples relative to the normal tissues, were clustered at chromosome 13q14, a region that is frequently deleted both in chronic lymphocytic leukemia (CLL), the most common human leukemia, and other cancers (55). Supporting such correlation, a large body of evidence indicate that over 50% of miRNA genes are positioned in cancer-associated genomic regions, such as fragile sites, minimal regions of loss of heterozygosity, minimal regions of amplification, or common breakpoint regions (56).Thanks to the development of high-throughput approaches, such as microarray and quantitative RT-PCR (qRT-PCR), global miRNA gene expression profiling has been successfully developed. Such analyses produced atlas of global miRNA expression patterns, in both cancers and normal tissues and definitely linked miRNAs to cancer. Following this discovery, significant changes in miRNA levels were also observed in several tumours without evident cytogenetic abnormalities, suggesting that the altered levels of miRNAs may be due to dysfunction in their biosynthesis or processing events. Such dysfunctions can result in the loss or gain of miRNA function, both contributing to initiation, progression and metastasis of human malignancies (57).

At the moment, it is current opinion that alterations in miRNA expression are the rule in human cancers and several studies assess that miRNAs can function as oncogenes or as tumour-suppressor genes in carcinogenesis (58). In particular, if a specific miRNA is over-expressed in primary tumours with respect to the normal tissue it may be considered a potential oncogene; on the contrary, if the miRNA is down-regulated in the tumoural tissue with respect to the normal one it may be regarded as a potential onco-suppressor gene. Notably, some miRNAs may act both as oncogenes and as tumour-suppressors, depending on the cellular context. This finding render more complex the functional classification of individual miRNAs, and makes essential to refer to specific tumours for describing their contribution to cancerogenesis.

The effect of the altered expression of miRNAs in a large number of malignancies, result in functional consequences by targeting main players in the eukaryotic survival, cell cycle and differentiation programs. Some miRNAs are deregulated in many cancers, suggesting that they may be involved in the control of basal processes as cell proliferation and apoptosis that are commonly deregulated in cancer; other miRNAs are, instead, deregulated in a tumour-specific manner according to their tissue-specificity.

6.1. Tumor suppressor and/or oncogenic miRNAs

Among the more studied miRNAs with tumour suppressor functions are the members of the let-7 family, which map to different human chromosomes; they are differentially expressed in different tissues and are involved in different tumours. let-7 is almost absent during developmental stages, whereas it is highly expressed in most differentiated tissues. Its oncosuppressive role has been largely studied in lung cancer where it is poorly expressed: it has been shown that let-7 directly controls cell proliferation by repressing the human RAS gene, which is overexpressed in lung cancer (59), and other cell cycle oncogenes as HMGA2 (54) and c-Myc (60). Among the most significant miRNAs with a putative role of oncogene, miR-21 that is overexpressed in almost all kinds of cancers, including colorectal cancer, pancreas endocrine and exocrine tumours, glioblastoma, ovary, lung. In breast carcinoma, miR-21 mediates cell survival and proliferation directly targeting the oncosuppressor genes PTEN, PDCD4 and TPM1 (61). Furthermore, it has been associated with advanced clinical stage and poor patient prognosis.

An interesting example of a miRNA that may function as oncosuppressor or oncogene, depending on the cellular context, is miR-125b, a homologue to lin-4 that is the first-identified miRNA. It has been shown that miR-125b contributes to the pathogenesis of prostate cancer displaying an oncogenic function. Shi and colleagues found that miR-125b is overexpressed in most clinical prostate cancer samples and in many cultured prostate cancer cell lines; accordingly, the ectopic expression of this miRNA stimulates androgen independent growth of prostate cancer cells, while repressing the expression of Bak1, a BCL2 family member, functioning as a proapoptotic regulator (62).

On the contrary, miR-125b exerts a tumour-suppressor function in breast cancer, where it is consistently down-regulated (61). About 25% of human breast cancers are associated with amplification and overexpression of the oncogenic receptor ERBB2; it has been demonstrated that the ectopic expression of miR-125b in a model breast cancer cell line directly repressed the expression of ERBB2, resulting in suppression of anchorage-dependent growth potential and inhibition of motility and invasive capabilities of breast cancer cells (63). Since increased expression of ERBB2 receptor was also reported in prostate cancer cell lines, the possibility that miR-125b modulates the expression of such gene in these tumour cells was investigated. Notably, miR-125b does not target ERBB2 gene in prostate cancer cells (62), indicating that the same miRNA may behave differently in different cancer cell types. miR-125b has also been described as a tumor-suppressive miRNA in neuroblastoma (NB) and medulloblastoma (MB). NB is a tumour of the sympathetic nervous system and represents the most frequent solid tumour of childhood, whereas MB is the most frequent brain malignancy in childhood. In both cancers miR-125b is down regulated and, consistently, its ectopic expression promotes cell growth arrest and apoptosis. This action is mediated by repressing the expression of the pro-proliferative truncated isoform of the neurotrophin receptor TrkC (t-TrkC), while the full length, pro-differentiative, TrkC isoform (fl-TrkC) is insensitive (64; 65).

miR-9 represents another example of a miRNA that may exert an oncogenic or oncosuppressive function, depending on the cell context. In breast cancer cells, miR-9 is upregulated and function as an oncogene by directly repressing the expression of E-cadherin; reduction of this key metastasis suppressing protein causes an increased cell motility and invasiveness (66; 67). A tumour growth-inhibitory function has been, instead, described for miR-9 in both NB and MB tumours, where this miRNA is down-regulated with respect to normal tissues. Accordingly its overexpression in NB and MB cell lines causes a decrease of cell proliferation and triggers neuronal differentiation (64; 65). miR-221 and miR-222 inhibit the growth of erythroblastic leukaemia while targeting the oncogene KIT and, therefore, functioning as tumour suppressors in erythroblastic cells (68). Differently, they may function as oncogenes by repressing tumour suppressors, as p27, p57, PTEN and TIMP3 in several human solid tumours (69).

6.2. miRNAs as tumor diagnostic tools

A very interesting aspect emerging from these studies is that different types of cancer display unique miRNA expression profiles that can be, therefore, used for tumour classification. This is a very important point since the first step towards the best treatment for cancer diseases is an early and accurate diagnosis. Surprisingly, the miRNomes are more informative than the mRNA profiling in predicting cancer type and stage. This is particularly true in the case of poorly differentiated tumours, the gene expression program of which is very different from that of their differentiated counterparts (70). Notably, it was assessed that the expression profiles of a small set of miRNAs may allow the classification of multiple cancers more accurately than data from about 16.000 mRNAs (71).

At the moment, several studies highlighted the relevance of miRNAs as tumour signatures that can be successfully employed as diagnostic markers for tumorigenesis. As an example, it was recently reported that a signature of only 13 miRNA is able to distinguish breast cancer and normal tissue with an accuracy of 100% (61). Furthermore, common aberrant miRNA exression profile allowed to distinguish both endocrine and acinar pancreatic cancers from normal pancreas. In particular, the expression of miR-103 and miR-107 together with contemporary lack of expression of miR-155 is diagnostic of tumours; moreover, a subset of 10 miRNAs distinguish endocrine from acinar tumours , whereas the overexpression of miR-204 of miR-21 correlate with insulinomas and liver metastasis respectively. This is an example of how a small set of miRNAs provide information on neoplastic transformation and progression of malignancy (72). A differential expression pattern of 32 miRNAs has been associated to the three neuroblastoma tumor subtypes (71). Recently, Ferretti and colleagues carried out, by high throughput analysis, the first miRNA expression profiling of human primary Medulloblastoma (MB). The authors showed that specific miRNA signatures may distinguish tumours from either adult or fetal normal tissues and that typical miRNA expression patterns allow the classification of MB histotypes, correlating with disease risk stratification (65). Through a genome-wide miRNA expression profiling in a large number of normal and tumour breast tissues the existence of a breast cancer-specific miRNA signature was also demonstrated (73).

7. EPIGENETIC CONTROL OF miRNA GENES

It is now becoming clear that miRNAs not only function in an epigenetic manner by post-transcriptionally regulating gene expression, but may also be targets of the epigenetic machinery (61). The study of epigenetic and transcriptional regulation of miRNA gene expression is, therefore, essential for a better understanding of their role in cancer.

Among the epigenetic mechanisms controlling miRNA gene expression, changes in DNA methylation pattern have been described and aberrant miRNA gene methylation has been associated to human tumorigenesis. In particular, DNA hypermethylation of potential oncosupressive miRNA genes may cause their downregulation; conversely, hypomethylation of miRNA genes with potential oncogenic function may cause their upregulation. In both cases the epigenetic regulation contributes to malignant transformation. It has been shown that half of miRNAs are associated to CpG islands, further suggesting that they may be targets for this kind of gene regulation (74). For instance, miR-127 is transcriptionally inactivated by CpG island hypermethylation (75). Human miR-127 is constitutively expressed in normal tissues, whereas it is repressed in cancer cells and downregulated in 75% of primary tumours, suggesting its potential oncosuppressive role. In particular, this miRNA, as part of miRNA cluster, is embedded within a CpG island and is highly induced (49-fold) after the simultaneous treatment with the chromatin-remodeling drugs 5-aza-2'-deoxycytidine and 4-phenylbutyric acid, which inhibit DNA methylation and histone deacetylase, respectively. This suggests that miR-127 is epigenetically silenced in cancer cells (75). Another example proving that DNA hypermethylation may contribute to the transcriptional downregulation of miRNAs in human tumours is miR-124a (76). This miRNA is embedded in a CpG island that is unmethylated in normal colon tissues, whereas it is hypermethylated in a collection of human cancer cell lines as well as in primary tumours from colon, breast, lung carcinomas, leukemias and limphomas; in colorectal tumours, miR-124a hypermethylation was observed in 75% of patients. RT-PCR analyses showed that mature miR-124a was absent in the wild type colon cancer HCT116 cell lines with hypermethylated CpG islands; notably miR-124a production was restored in the same cell line after disruption of methyltransferases (DNMT1 and DNMT3b) as well as after treatment with the DNA-demethylating agents. These results highlight the association between miR-124a methylation and loss of miRNA expression and clearly demonstrate that in cancer cells miR-124 is locked in a transcriptionally inactive state (76). Furthermore, aberrant hypermethylation was observed for miR-9-1, miR-124a-3, miR-148, miR-152 and miR-663 in 34-86% of cases in a series of 71 primary human breast cancer specimens (77). In particular, for miR-9-1 a direct correlation between methylation level and reduction of expression was demonstrated in a subset of primary human breast cancers. Also miR-34b and miR-34c, that are clustered, are downmodulated in colon cancer following DNA hypermethylation (78). Among the miRNA genes that are hypomethylated in cancer, let-7a-3, that belongs to the archetypal family of let-7 miRNA genes. The human let-7a-3 gene, with putative oncogenic function, is associated with a CpG island and is highly methylated in normal tissues but hypomethylated in some lung adenocarcinomas. Such DNA hypomethylation facilitates epigenetic activation of the gene, and the following elevated expression level of let-7a-3 in a human lung cancer cell line results in enhanced tumour phenotypes, and oncogenic changes in transcription profiles (79). Recently, Tsai and colleagues described the epigenetic control of the expression of a primate-specific miRNA cluster (C19MC), which spans about 100kb on human chromosome 19 and comprises 46 miRNAs. Such miRNA cluster is specifically expressed in the placenta, and its expression pattern is associated with methylation state of a distal CpG-rich region located at 17.6 kb upstream; in other cell types, where the miRNA cluster is poorly expressed, it is silenced by hypermethylation of the distal CpG-rich region (80).

However, methylation is not the only epigenetic mechanism that influences miRNA gene expression. In fact, also histone acetylation has been involved, as revealed by alteration of miRNA expression levels after inhibition of histone deacetylase. In this regard, Scott and colleagues explored the changes in miRNA profiles after treatment of SKBr3 breast carcinoma cells with the potent hydroxamic acid HDAC inhibitor (LAQ824). They found that the expression level of about 30 miRNAs was altered, with 22 miRNAs down regulated and 5 miRNAs upregulated (63).

8. miRNA CONTROL OF EPIGENETIC MECHANISMS

If on one hand miRNAs may be targets of the epigenetic machinery, on the other hand they may control epigenetic machinery, directly targeting its enzyme components. Such reciprocal action defines a highly controlled feed-back mechanism that contributes to the regulation of the sophisticated networks underlying the maintenance of a non pathological state.

Among these miRNAs, called "epi-miRNAs" (81), miR-290 cluster, the loss of which in Dicer-1-deficient mouse embryonic stem cells causes downregulation of the methyltransferases DNMT3a, DNMT3b and DNMT1. Such effect is mediated by upregulation of their repressor retinoblastoma-like 2 protein (RbI2), which is a direct target of miR-290 (82; 83). The consequent aberrant DNA methylation impairs the embryonic stem cell differentiation program (82) and affected the length of the hypomethylated telomeres (83). Other epi-miRNAs affecting DNA methylation have been described. Among them, miR-29 family that directly regulates the "de novo" DNA methyltransferases DNMT-3A and 3B (84) and indirectly controls the maintenance DNA methyltransferase DNMT1 (85). Such enzyme is the target of other miRNA as miR-148a, miR-152 and miR301 in cholangiocarcinoma (86). Another miRNA linked to the epigenetic machinery is miR-137, which is involved in proliferation and differentiation in vitro and in vivo of adult neural stem cells. Overexpression of this miRNA promotes proliferation, whereas reduction of its level induces neuronal cell differentiation. Notably, miR-137 is epigenetically controlled by MeCP2, a DNA methyl-CpG-binding protein, and in turn represses the expression of EZH2, a histone methyltransferase and catalytic subunit of the polycomb repressive complex 2 (PRC2). Repression of Ezh2 results in a global reduction of histone H3 trimethyl lysine 27, (Figure 1). (87).

EZH2 is also a direct target of miR-101;the abnormal downregulation of this miRNA in several tumours is associated to the overexpression of EZH2, frequently observed in cancer. (88). Finally, some miRNAs controlling enzymes directing histone acetylation have been reported. Among them, miR-1 and miR140 directly target HDAC4 (89), whereas miR-449a controls HDAC1 (90).

9. SUMMARY AND PERSPECTIVES

The expression of the genome is mirrored in the transcriptome and proteome profiling and specify the fate of each cell type.The epigenetic regulation impact on the chromatin dynamic and determine the overall compaction and subsequent gene expression. Nevertheless, the extent to which epigenetic changes are heritable or if and how post-translational modifications pass on the memory of a given chromatin state to the progeny is still debated. It has been proposed that at least persistent histone marks, like H3K9me, are able to be transmitted through cell cycles and therefore represent a canonical heritable trait. Notably, epigenetic factors and miRNAs are tightly integrated in complex regulatory circuitries that underpin differentiation process. The reciprocal and concerted action between epigenetic machinery and proper expression of miRNAs may create regulatory loops that can simultaneously control hundreds of target genes, thus ensuring the harmonized gene expression and execution of differentiation programs (Figure 2). Disruption of this critical cross regulation may lead to cell transformation and cancer. A matter of future investigation and scientific challenge will be to fully dissect the close reciprocal interactions and the circuitries which cross regulate miRNAs and epigenetic mechanisms. This field of molecular biology may reveal novel precocious diagnostic tools or powerful molecules able to block uncontrolled cell proliferation and drive cell fate toward differentiative pathways.

10. AKNOWLEDGEMENTS

We apologise for the Authors not cited in this review due to space limitations. This work was supported by PRIN 20075h7a9_003. European Union project SIROCCO (LSHG-CT-2006-037900) and Istituto Italiano di Tecnologia SEED-project.

11. REFERENCES

1. Shelley L. Berger, Tony Kouzarides, Ramin Shiekhattar and Ali Shilatifard. An operational definition of epigenetics. Genes Dev. 23, 781-783. (2009)
doi:10.1101/gad.1787609


2. Eileen Gibney and Catherine Nolan. Epigenetics and gene expression. Heredity 105, 4-13. (2010)
doi:10.1038/hdy.2010.54

3. Frank Sleutels, Ronald Zwart and Denise Barlow. The non-coding Air RNA is required for silencing autosomal imprinted genes. Nature 415, 810-3. (2002)


4. Takashi Nagano, Jennifer A. Mitchell, Lionel A. Sanz, Florian M. Pauler, Anne C. Ferguson-Smith, Robert Feil, Peter Fraser. The Air noncoding RNA epigenetically silences transcription by targeting G9a to chromatin. Science 322, 1717-1720. (2008)
doi:10.1126/science.1163802

5. Ahmad M. Khalil, Mitchell Guttman, Maite Huarte, Manuel Garber, Arjun Raj, Dianali Rivea Morales, Kelly Thomas, Aviva Presser, Bradley E. Bernstein, Alexander van Oudenaarden, Aviv Regev, Eric S. Lander and John L. Rinn. Many human large intergenic noncoding RNAs associate with chromatin-modifying complexes and affect gene expression. Proc Nat Acad Sci USA 106, 11667-11672. (2009)
doi:10.1073/pnas.0904715106

6. Shikhar Sharma, Theresa Kelly and Peter Jones. Epigenetics in cancer. Carcinog. 31, 27-36. (2010)
doi:10.1093/carcin/bgp220


7. Masahiro Kaneda, Masaki Okano, Kenichiro Hata, Takashi Sado, Naomi Tsujimoto, En Li and Hiroyuki Sasaki. Essential role for de novo DNA methyltransferase Dnmt3a in paternal and maternal imprinting. Nature 429, 900-903. (2004)
doi:10.1038/nature02633

8. Linda J. Ball, Natalia V. Murzina, R.William Broadhurst, Andrew R.C. Raine, Sharon J. Archer, Francesca J. Stott, Alexey G. Murzin, Prim B. Singh, Peter J. Domaille and Ernest D. Laue. Structure of the chromatin binding (chromo) domain from mouse modifier protein 1. EMBO J. 16, 2473-2481. (1997)
doi:10.1093/emboj/16.9.2473

9. Peter R. Nielsen, Daniel Nietlispach, Helen R. Mott, Juliana Callaghan, Andrew Bannister, Tony Kouzarides, Alexey G. Murzin, Natalia V. Murzina and Ernest D. Laue. Structure of the HP1 chromodomain bound to histone H3 methylated at lysine 9. Nature 416, 103-107. (2002)
doi:10.1038/nature722

10.Xinsheng Nan, Huck-Hui Ng, Colin A. Johnson, Carol D. Laherty, Bryan M. Turner, Robert N. Eisenman and Adrian Bird. Transcriptional repression by the methyl-CpG-binding protein MeCP2 involves a histone deacetylase complex. Nature 393, 386-389. (1998)
doi:10.1038/30764

11.Peter A Jones and Peter W Laird, Cancer epigenetics comes of age. Nat. Genet. 21, 163-167. (1999)
doi:10.1038/5947

12.Andrew P. Feinberg, Rolf Ohlsson and Steven Henikoff. The epigenetic progenitor origin of human cancer. Nat. Rev. Genet. 7, 21-33. (2006)
doi:10.1038/nrg1748

13.Yong Wang and Frederick C.C. Leung. An evaluation of new criteria for CpG islands in the human genome as gene markers. Bioinform, 20, 1170-1177. (2004)
doi:10.1093/bioinformatics/bth059

14.Miho M. Suzuki and Adrian Bird. DNA methylation landscapes: provocative insights from epigenomics, Nat. Rev. Genet. 9, 465-476. (2008)
doi:10.1038/nrg2341


15.Franáois Fuks, Paul J. Hurd, Daniel Wolf, Xinsheng Nan, Adrian P. Bird and Tony Kouzarides. The methyl-CpG-binding protein MeCP2 links DNA methylation to histone methylation. J. Biol, Chem. 278, 4035-4040. (2003)
doi:10.1074/jbc.M210256200

16. Ann S. Wilson, Barbara E. Power, and Peter L. Molloy. DNA hypomethylation and human diseases. Biochim. Biophys. Acta. 1775, 138-62. (2007)
doi:10.1016/j.bbcan.2006.08.007

17.Eric Yuan, Chi-Ming Li, Darrell J. Yamashiro, Jessica Kandel, Harshwardhan Thaker, Vundavalli V. Murty and Benjamin Tycko. Genomic profiling maps loss of heterozygosity and defines the timing and stage dependence of epigenetic and genetic events in Wilms' tumors. Mol Cancer Res. 9, 493-502. (2005)
doi:10.1158/1541-7786.MCR-05-0082

18. Greger V, Passarge E, Hîpping W, Messmer E, Horsthemke B. Epigenetic changes may contribute to the formation and spontaneous regression of retinoblastoma. Hum. Genet. (1989) 83:155-158.
doi:10.1007/BF00286709

19.Stephen B Baylin. DNA methylation and gene silencing in cancer. Nat. Clin. Pract. Oncol. 2, S4-S11. (2005).
doi:10.1038/ncponc0354

20.Chaozhong Long, Bangliang Yin, Qianjin Lu, Xinmin Zhou, Jianguo Hu, Yifeng Yang. Fenglei Yu, and Yunchang Yuan. Promoter hypermethylation of the RUNX3 gene in esophageal squamous cell carcinoma. Cancer Invest. 25, 685-690. (2007)
doi:10.1080/07357900701561131

21.Yoshimitsu Akiyama, Neil Watkins, Hiromu Suzuki, Kam-Wing Jair, Manon van Engeland, Manel Esteller, Hidekazu Sakai, Chun-Yan Ren, Yasuhito Yuasa, James G. Herman and Stephen B. Baylin. GATA-4 and GATA-5 transcription factor genes and potential downstream antitumor target genes are epigenetically silenced in colorectal and gastric cancer. Mol. Cell. Biol. 23, 8429-8439. (2003)
doi:10.1128/MCB.23.23.8429-8439.2003

22.Luciano Di Croce,Veronica A. Raker, Massimo Corsaro, Francesco Fazi, Mirco Fanelli, Mario Faretta, Francois Fuks, Francesco Lo Coco, Tony Kouzarides, Clara Nervi, Saverio Minucci, Pier Giuseppe Pelicci. Methyltransferase recruitment and DNA hypermethylation of target promoters by an oncogenic transcription factor. Science 295, 1079-82. (2002)
doi:10.1126/science.1065173

23. Daniel Weisenberger, Kimberly Siegmund, Mihaela Campan, Joanne Young, Tiffany Long, Mark Faasse, Gyeong Hoon Kang, Martin Widschwendter, Deborah Weener, Daniel Buchanan, Hoey Koh, Lisa Simms, Melissa Barker, Barbara Leggett, Joan Levine, Myungjin Kim, Amy French, Stephen Thibodeau, Jeremy Jass, Robert Haile and Peter Laird. CpG island methylator phenotype underlies sporadic microsatellite instability and is tightly associated with BRAF mutation in colorectal cancer. Nat. Genet. 38, 787-793. (2006).
doi:10.1038/ng1834

24.Karolin Luger, Armin W. MÑder, Robin K. Richmond, David F. Sargent and Timothy J. Richmond. Crystal structure of the nucleosome core particle at 2.8 A resolution. Nature 389, 251-260. (1997)
doi:10.1038/38444

25. C. David Allis, Shelley L. Berger, Jacques Cot, Sharon Dent, Thomas Jenuwien, Tony Kouzarides, Lorraine Pillus, Danny Reinberg, Yang Sh, Ramin Shiekhattar, Ali Shilatifard, Jerry Workman and Yi Zhang. New nomenclature for chromatin-modifying enzymes. Cell. 131, 633-6. (2007)

doi:10.1016/j.cell.2007.10.039

26.Sean D Taverna, Haitao Li, Alexander J Ruthenburg, C David Allis and Dinshaw J Patel. How chromatin-binding modules interpret histone modifications: lessons from professional pocket pickers. Nat Struct Mol Biol. 14, 1025-1040. (2007)
doi:10.1038/nsmb1338

27.Mary Grace Goll and Timothy H. Bestor Histone modification and replacement in chromatin activation, Genes Dev. 16,1739-1742. (2002)
doi:10.1101/gad.1013902

28.Thomas Jenuwein and C. David Allis. Translating the histone code. Science 293, 1074-1080. (2001)
doi:10.1126/science.1063127

29.John A Latham and Sharon Y R Dent. Cross-regulation of histone modifications. Nat. Struct. Mol. Biol. 14, 1017-1024. (2007)
doi:10.1038/nsmb1307

30.Bannister A.J., R. Schneider, F.A. Myers, A.W. Thorne, C. Crane-Robinson and T. Kouzarides, Spatial distribution of di- and tri-methyl lysine 36 of histone H3 at active genes. J. Biol. Chem. 280, 17732-17736. (2005)
doi:10.1074/jbc.M500796200

31.Tony Kouzarides. Chromatin modifications and their function. Cell. 128. 693-705. (2007)
doi:10.1016/j.cell.2007.02.005


32.Antoine H.F.M. Peters, Jacqueline E. Mermoud, Donal O'Carroll, Michaela Pagani, Dieter Schweizer, Neil Brockdorff and Thomas Jenuwein. Histone H3 lysine 9 methylation is an epigenetic imprint of facultative heterochromatin. Nat. Genet. 30, 77-80. (2002)
doi:10.1038/ng789

33.Bing Li, Madelaine Gogol, Mike Carey, Daeyoup Lee, Chris Seidel, Jerry L. Workman. Combined action of PHD and chromo domains directs the Rpd3S HDAC to transcribed chromatin. Science. 316, 1050-1054. (2007)
doi:10.1126/science.1139004

34.Joseph B. Rayman, Yasuhiko Takahashi, Vahan B. Indjeian, Jan-Hermen Dannenberg, Steven Catchpole, Roger J. Watson, Hein te Riele, and Brian David Dynlacht. E2F mediates cell cycle-dependent transcriptional repression in vivo by recruitment of an HDAC1/mSin3B corepressor complex. Genes Dev. 16, 933-947. (2002)
doi: 10.1101/gad.969202.

35.Peter Jones and Stephen Baylin. The epigenomics of cancer. Cell, 128, 683-692. (2007)

doi:10.1016/j.cell.2007.01.029

36.Sam Thiagalingam, Kuang-Hung Cheng, Hiunjoo Lee, Nora Mineva, Arunthathi Thiagalingam and Jose Ponte. Histone deacetylases: unique players in shaping the epigenetic histone code. Ann. N. Y. Acad. Sci. 983, 84-100. (2003)
doi:10.1111/j.1749-6632.2003.tb05964.x

37.Manel Esteller. Aberrant DNA methylation as a cancer-inducing mechanism, Annu. Rev. Pharmacol. Toxicol. 45, 629-656. (2005)
doi:10.1146/annurev.pharmtox.45.120403.095832

3 David B. Seligson, Steve Horvath, Tao Shi, Hong Yu, Sheila Tze, Michael Grunstein and Siavash K. Kurdistani. Global histone modification patterns predict risk of prostate cancer recurrence. Nature. 435, 1262-6. (2005)
doi:10.1038/nature03672

39.Krivtsov A.V. and S.A. Armstrong, MLL translocations, histone modifications and leukaemia stem-cell development, Nat. Rev. Cancer. 7, 823-833. (2007)
doi:10.1038/nrc2253

40. Hein Visser, Marco Gunster, Hanneke Kluin-Nelemans, Erik Manders, Frank Raaphorst, Chris Meijer, Roel Willemze and Arie Otte. The Polycomb group protein EZH2 is upregulated in proliferating, cultured human mantle cell lymphoma. Br J Haematol. 112, 950-8. (2001)
doi:10.1046/j.1365-2141.2001.02641.x

41. Celina Kleer, Qi Cao, Sooryanarayana Varambally, Ronglai Shen, Ichiro Ota, Scott Tomlins, Debashis Ghosh, Richard Sewalt, Arie Otte, Daniel Hayes, Michael Sabel, Donna Vivant, Stephen Weiss, Mark Rubin and Arul Chinnaiyan. EZH2 is a marker of aggressive breast cancer and promotes neoplastic transformation of breast epithelial cells, Proc. Natl. Acad. Sci. U.S.A. 100,11606-11611. (2003)
doi:10.1073/pnas.1933744100


42.Daniel Rhodes, Martin Sanda, Arie Otte, Arul Chinnaiyan, Mark Rubin. Multiplex biomarker approach for determining risk of prostate-specific antigen-defined recurrence of prostate cancer, J. Natl. Cancer Inst. 95, 661-668.(2003)
doi:10.1093/jnci/95.9.661

43.Raffaella Villa, Diego Pasini, Arantxa Gutierrez, Lluis Morey, Manuela Occhionorelli, Emmanuelle Vire, Josep F. Nomdedeu, Thomas Jenuwein, Pier Giuseppe Pelicci, Saverio Minucci, Francois Fuks, Kristian Helin and Luciano Di Croce. Role of the polycomb repressive complex 2 in acute promyelocytic leukemia. Cancer Cell. 6, 513-525. (2007)
doi:10.1016/j.ccr.2007.04.009

44.Leenders GJ, Dukers D, Hessels D, van den Kieboom SW, Hulsbergen CA, Witjes JA, Otte AP, Meijer CJ, Raaphorst FM. Polycomb-group oncogenes EZH2, BMI1, and RING1 are overexpressed in prostate cancer with adverse pathologic and clinical features, Eur. Urol. 52, 455-463. (2007)
doi:10.1016/j.eururo.2006.11.020

45.Kitabayashi I, Y. Aikawa, A. Yokoyama, F. Hosoda, M. Nagai, N. Kakazu, T. Abe and M. Ohki, Fusion of MOZ and p300 histone acetyltransferases in acute monocytic leukemia with a t(8;22)(p11;q13) chromosome translocation, Leukem. 15, 89-94. (2001)
doi:10.1038/sj.leu.2401983


46. Myoung Sook Kim, Ho Jeong Kwon, You Mie Lee, Jin Hyen Baek, Jae-Eun Jang, Sae-Won Lee, Eun-Joung Moon, Hae-Sun Kim, Seok-Ki Lee, Hae Young Chung, Chul Woo Kim and Kyu-Won Kim. Histone deacetylases induce angiogenesis by negative regulation of tumor suppressor genes, Nat. Med. 7, 437-443. (2001)
doi:10.1038/86507

47. Ananya Manuyakorn, Rebecca Paulus, James Farrell, Nicole Dawson, Sheila Tze, Gardenia Cheung-Lau, Oscar Joe Hines, Howard Reber, David Seligson, Steve Horvath, Siavash Kurdistani, Chandhan Guha and David Dawson. Cellular histone modification patterns predict prognosis and treatment response in resectable pancreatic adenocarcinoma: results from RTOG 9704. J Clin Oncol. 28, 1358-1365. (2010)
doi:10.1200/JCO.2009.24.5639


48.Elsheikh SE, Green AR, Rakha EA, Powe DG, Ahmed RA, Collins HM, Soria D, Garibaldi JM, Paish CE, Ammar AA, Grainge MJ, Ball GR, Abdelghany MK, Martinez Pomares L, Heery DM, Ellis . Global histone modifications in breast cancer correlate with tumor phenotypes, prognostic factors, and patient outcome. Cancer Res. 69, 3802-3809. (2009)
doi:10.1158/0008-5472.CAN-08-3907

49. Mario F Fraga, Esteban Ballestar, Ana Villar-Garea, Manuel Boix-Chornet, Jesus Espada, Gunnar Schotta, Tiziana Bonaldi, Claire Haydon, Santiago Ropero, Kevin Petrie, N Gopalakrishna Iyer, Alberto Pérez-Rosado, Enrique Calvo, Juan A Lopez, Amparo Cano, Maria J Calasanz, Dolors Colomer, Miguel Angel Piris, Natalie Ahn, Axel Imhof, Carlos Caldas, Thomas Jenuwein and Manel Esteller. Loss of acetylation at Lys16 and trimethylation at Lys20 of histone H4 is a common hallmark of human cancer. Nat Genet. 37, 391-400. (2005)


50.Daehyun Baek, Judit Villen, Chanseok Shin, Fernando D. Camargo, Steven P. Gygi and David P. Bartel. The impact of microRNAs on protein output. Nature 455, 64-71. (2008)
doi:10.1038/nature07242

51.Matthias Selbach, Bjîrn SchwanhÑusser, Nadine Thierfelder, Zhuo Fang, Raya Khanin and Nikolaus Rajewsky. Widespread changes in protein synthesis induced by microRNAs. Nature. 455, 58-63. (2008)
doi:10.1038/nature07228

52.Isaac Bentwich, Amir Avniel, Yael Karov, Ranit Aharonov, Shlomit Gilad, Omer Barad, Adi Barzilai, Paz Einat, Uri Einav, Eti Meiri, Eilon Sharon, Yael Spector and Zvi Bentwich. Identification of hundreds of conserved and nonconserved human microRNAs. Nat. Genet. 37, 766-770. (2005)
doi:10.1038/ng1590

53.Eugene Berezikov, Victor Guryev, JosÇ van de Belt, Erno Wienholds, Ronald H.A. Plasterk and Edwin Cuppen. Phylogenetic shadowing and computational identification of human microRNA genes. Cell 120, 21-24. (2005)
doi:10.1016/j.cell.2004.12.031

54. Yong Sun Lee, Anindya Dutta. MicroRNA in cancer. Annu. Rev. Mech. Dis. 4, 199-227. (2009)
doi:10.1146/annurev.pathol.4.110807.092222

55.George Adrian Calin, Calin Dan Dumitru, Masayoshi Shimizu, Roberta Bichi, Simona Zupo, Evan Noch, Hansjuerg Aldler, Sashi Rattan, Michael Keating, Kanti Rai, Laura Rassenti, Thomas Kipps, Massimo Negrini, Florencia Bullrich, and Carlo M. Croce. Frequent deletions and down-regulation of micro- RNA genes miR15 and miR16 at 13q14 in chronic lymphocytic leukemia. Proc Natl Acad Sci U S A. 99, 15524-15529. (2002)
doi:10.1073/pnas.242606799

56.George Adrian Calin, Cinzia Sevignani, Calin Dan Dumitru, Terry Hyslop, Evan Noch, Sai Yendamuri, Masayoshi Shimizu, Sashi Rattan, Florencia Bullrich, Massimo Negrini and Carlo M. Croce. Human microRNA genes are frequently located at fragile sites and genomic regions involved in cancers. Proc Natl Acad Sci U S A. 101, 2999-3004. (2004)
doi:10.1073/pnas.0307323101

57. Lin He, Michael Thomson, Michael Hemann, Eva Hernando-Monge, David Mu, Summer Goodson, Scott Powers, Carlos Cordon-Cardo, Scott Lowe, Gregory Hannon and Scott Hammond. A microRNA polycistron as a potential human oncogene. Nature. 435, 828-833. (2005)
doi:10.1038/nature03552


58.Croce Carlo Maria. Causes and consequences of microRNA dysregulation in cancer. Nat Rev Genet.10, 704-714. (2009)
doi:10.1038/nrg2634

59.Steven Johnson, Helge Grosshans, Jaclyn Shingara, Mike Byrom, Rich Jarvis, Angie Cheng, Emmanuel Labourier, Kristy Reinert, David Brown and Frank Slack. RAS is regulated by the let-7 microRNA family. Cell. 120, 635-47. (2005)
doi:10.1016/j.cell.2005.01.014

60.Valerie B. Sampson, Nancy H. Rong, Jian Han, Qunying Yang, Virginie Aris, Patricia Soteropoulos, Nicholas J. Petrelli, Stephen P. Dunn and Leslie J. Krueger. MicroRNA let-7a down-regulates MYC and reverts MYC-induced growth in Burkitt lymphoma cells. Cancer Res. 67, 9762-9770. (2007)
doi:10.1158/0008-5472.CAN-07-2462


61.Marilena V. Iorio, Carlo M. Croce. MicroRNAs in cancer: small molecules with a huge impact. J Clin Oncol. 27, 5848-5856. (2009)
doi:10.1200/JCO.2009.24.0317

62.Xu-Bao Shi, Lingru Xue, Joy Yang, Ai-Hong Ma, Jianjun Zhao, Ma Xu, Clifford G. Tepper, Christopher P. Evans, Hsing-Jien Kung, and Ralph W. deVere White. An androgen-regulated miRNA suppresses Bak1 expression and induces androgen-independent growth of prostate cancer cells. Proc Natl Acad Sci U S A. 104, 19983-19988. (2007)
doi:10.1073/pnas.0706641104

63.Gary K. Scott, Michael D. Mattie, Crystal E. Berger, Stephen C. Benz, and Christopher C. Benz. Rapid alteration of microRNA levels by histone deacetylase inhibition. Cancer Res. 66, 1277-1281. (2006)
doi:10.1158/0008-5472.CAN-05-3632

64.Laneve Pietro, Di Marcotullio Lucia, Gioia Ubaldo, Fiori Micol E, Ferretti Elisabetta, Gulino Alberto, Bozzoni Irene, Caffarelli Elisa.The interplay between microRNAs and the neurotrophin receptor tropomyosin-related kinase C controls proliferation of human neuroblastoma cells. Proc Natl Acad Sci U S A.104, 7957-7962. (2007)
doi:10.1073/pnas.0700071104

65.Elisabetta Ferretti, Enrico De Smaele, Agnese Po, Lucia Di Marcotullio, Emanuele Tosi, Maria Salome B. Espinola, Concezio Di Rocco, Riccardo Riccardi, Felice Giangaspero, Alessio Farcomeni, Italo Nofroni, Pietro Laneve, Ubaldo Gioia, Elisa Caffarelli, Irene Bozzoni, Isabella Screpanti, Alberto Gulino. MicroRNA profiling in human medulloblastoma. Int J Cancer.124, 568-577. (2009)
doi:10.1002/ijc.23948

66. Li Ma, Jennifer Young, Harsha Prabhala, Elizabeth Pan, Pieter Mestdagh, Daniel Muth, Julie Teruya-Feldstein, Ferenc Reinhardt, Tamer Onder, Scott Valastyan, Frank Westermann, Frank Speleman, Jo Vandesompele and Robert Weinberg. miR-9, a MYC/MYCN-activated microRNA, regulates E-cadherin and cancer metastasis. Nat Cell Biol. 12, 247-56. (2010)



67.Yeesim Khew-Goodall and Gregory J. Goodall. Myc-modulated miR-9 makes more metastases. Nat Cell Biol. 12,209-11.


68.Nadia Felli, Laura Fontana, Elvira Pelosi, Rosanna Botta, Desire Bonci, Francesco Facchiano,Francesca Liuzzi, Valentina Lulli, Ornella Morsilli, Simona Santoro, Mauro Valtieri, George Adrian Calin, Chang-Gong Liu, Antonio Sorrentino, Carlo M. Croce and Cesare Peschle. microRNAs 221 and 222 inhibit normal erythropoieis and erythroleukemic cell growth via kit receptor down-moulation. Proc Natl Acad Sci USA 102, 18081-18086. (2005)
doi:10.1073/pnas.0506216102

69.Marilena V. Iorio, Manuela Ferracin, Chang-Gong Liu, Angelo Veronese, Riccardo Spizzo, Silvia Sabbioni, Eros Magri, Massimo Pedriali, Muller Fabbri, Manuela Campiglio, Sylvie MÇnard, Juan P. Palazzo, Anne Rosenberg, Musini, Stefano Volinia, Italo Nenci, George A. Calin, Patrizia Querzoli, Massimo Negrini, and Carlo M. Croce. microRNA gene expression deregulation in human breast cancer. Cancer Res. 65, 7065-7070. (2005)
doi:10.1158/0008-5472.CAN-05-1783

70.Sridhar Ramaswamy, Pablo Tamayo, Ryan Rifkin, Sayan Mukherjee, Chen-Hsiang Yeang, Michael Angelo, Christine Ladd, Michael Reich, Eva Latulippe, Jill P. Mesirov, Tomaso Poggio, William Gerald, Massimo Loda, Eric S. Lander, and Todd R. Golub. Proc Natl Acad Sci U S A. 98,15149-15154. (2001)
doi:10.1073/pnas.211566398

71.Jun Lu, Gad Getz, Eric A. Miska, Ezequiel Alvarez-Saavedra, Justin Lamb, David Peck, Alejandro Sweet-Cordero, Benjamin L. Ebert, Raymond H. Mak, Adolfo A. Ferrando, James R. Downing, Tyler Jacks, H. Robert Horvitz and Todd R. Golub. MicroRNA expression profiles classify human cancers. Nature. 435, 834-838. (2005)
doi:10.1038/nature03702

72.Claudia Roldo, Edoardo Missiaglia, John P. Hagan, Massimo Falconi, Paola Capelli, Samantha Bersani, George Adrian Calin, Stefano Volinia, Chang-Gong Liu, Aldo Scarpa, Carlo M. Croce. MicroRNA expression abnormalities in pancreatic endocrine and acinar tumors are associated with distinctive pathologic features and clinical behavior. J Clin Oncol. 24, 4677-4684. (2006)
doi:10.1200/JCO.2005.05.5194

73.Marilena V. Iorio, Rosa Visone, Gianpiero Di Leva, Valentina Donati, Fabio Petrocca, Patrizia Casalini, Cristian Taccioli, Stefano Volinia, Chang-Gong Liu, Hansjuerg Alder, George A. Calin, Sylvie MÇnard, and Carlo M. Croce. MicroRNA signatures in human ovarian cancer. Cancer Res. 67, 8699-8707. (2007)
doi:10.1158/0008-5472.CAN-07-1936

74.Barbara Weber, Carlo Stresemann, Bodo Brueckner and Frank Lyko. Methylation of Human MicroRNA Genes in Normal and Neoplastic Cells. Cell Cycle 6,1001-1005. (2007)


75.Yoshimasa Saito, Gangning Liang, Gerda Egger, Jeffrey M. Friedman, Jody C. Chuang, Gerhard A. Coetzee and Peter A. Jones. Specific activation of microRNA-127 with downregulation of the proto-oncogene BCL6 by chromatin-modifying drugs in human cancer cells. Cancer Cell. 9,435-443. (2006)
doi:10.1016/j.ccr.2006.04.020

76.Amaia Lujambio, Santiago Ropero, Esteban Ballestar, Mario F. Fraga, Celia Cerrato, Fernando SetiÇn, Sara Casado, Ana Suarez-Gauthier, Montserrat Sanchez-Cespedes, Anna Gitt, Inmaculada Spiteri, Partha P. Das, Carlos Caldas, Eric Miska, and Manel Esteller. Genetic Unmasking of an Epigenetically Silenced microRNA in Human Cancer Cells. Cancer Res. 67,1424-1429. (2007)
doi:10.1158/0008-5472.CAN-06-4218


77.U Lehmann, B Hasemeier, M Christgen, M Muller, D Romermann, F Langer and H Kreipe. Epigenetic inactivation of microRNA gene hsa-mir-9-1 in human breast cancer. J Pathol. 214, 17-24. (2008)
doi:10.1002/path.2251

78.Minoru Toyota, Hiromu Suzuki, Yasushi Sasaki, Reo Maruyama, Kohzoh Imai, Yasuhisa Shinomura, and Takashi Tokino. Epigenetic silencing of microRNA-34b/c and B-cell translocation gene 4 is associated with CpG island methylation in colorectal cancer. Cancer Res. 68, 4123-4132. (2008)

79.Bodo Brueckner, Carlo Stresemann, Ruprecht Kuner, Cora Mund, Tanja Musch, Michael Meister, Holger SÅltmann, and Frank Lyko. The human let-7a-3 locus contains an epigenetically regulated microRNA gene with oncogenic function. Cancer Res. 67, 1419-1423. (2007)

80.Kuo-Wang Tsai, Hsiao-Wei Kao, Hua-Chien Chen, Su-Jen Chen and Wen-chang Lin.Epigenetic control of the expression of a primate-specific microRNA cluster in human cancer cells. Epigenetics 4, 587-592. (2009)
doi:10.4161/epi.4.8.10230

81.Marilena V. Iorio, Claudia Piovan, and Carlo M. Croce. Interplay between microRNAs and the epigenetic machinery: An intricate network. Biochim Biophys Acta. in press (2010)

doi:10.1016/j.bbagrm.2010.05.005

82.Roberta Benetti, Susana Gonzalo, Isabel Jaco, Purificacion Munoz, Susana Gonzalez, Stefan Schoeftner, Elizabeth Murchison, Thomas Andl, Taiping Chen, Peter Klatt, En Li, Manuel Serrano, Sarah Millar, Gregory Hannon and Maria A Blasco. A mammalian microRNA cluster controls DNA methylation and telomere recombination via Rbl2-dependent regulation of DNA methyltransferases. Nat Struct Mol Biol. 15, 268-279. (2008)
doi:10.1038/nsmb.1399

83.Lasse Sinkkonen, Tabea Hugenschmidt, Philipp Berninger, Dimos Gaidatzis, Fabio Mohn, Caroline G Artus-Revel, Mihaela Zavolan, Petr Svoboda and Witold Filipowicz. MicroRNAs control de novo DNA methylation through regulation of transcriptional repressors in mouse embryonic stem cells. Nat Struct Mol Biol. 15, 259-267. (2008)
doi:10.1038/nsmb.1391

84.Muller Fabbri, Ramiro Garzon, Amelia Cimmino, Zhongfa Liu, Nicola Zanesi, Elisa Callegari, Shujun Liu, Hansjuerg Alder, Stefan Costinean, Cecilia Fernandez-Cymering, Stefano Volinia, Gulnur Guler, Carl D. Morrison, Kenneth K. Chan, Guido Marcucci, George A. Calin, Kay Huebner, and Carlo M. Croce. MicroRNA-29 family reverts aberrant methylation in lung cancer by targeting DNA methyltransferases 3A and 3B. Proc Natl Acad Sci U S A. 104, 15805-15810. (2007)
doi:10.1073/pnas.0707628104

85.Ramiro Garzon, Stefano Volinia, Chang-Gong Liu, Cecilia Fernandez-Cymering, Tiziana Palumbo, Flavia Pichiorri, Muller Fabbri, Kevin Coombes, Hansjuerg Alder, Tatsuya Nakamura, Neal Flomenberg, Guido Marcucci, George A. Calin, Steven M. Kornblau, Hagop Kantarjian, Clara D. Bloomfield, Michael Andreeff, and Carlo M. Croce MicroRNA signatures associated with cytogenetics and prognosis in acute myeloid leukemia. Blood. 111, 3183-3189. (2008)
doi:10.1182/blood-2007-07-098749

86.Chiara Braconi, Nianyuan Huang, and Tushar Patel. MicroRNA-dependent regulation of DNA methyltransferase-1 and tumor suppressor gene expression by interleukin-6 in human malignant cholangiocytes. Hepatology 51, 881-890. (2010)



87.Keith E. Szulwach, Xuekun Li, Richard D. Smrt, Yujing Li, Yuping Luo, Li Lin, Nicholas J. Santistevan, Wendi Li, Xinyu Zhao, and Peng Jin. Cross talk between microRNA and epigenetic regulation in adult neurogenesis. J Cell Biol. 189, 127-141. (2010)
doi:10.1083/jcb.200908151

88.Jeffrey M. Friedman, Gangning Liang, Chun-Chi Liu, Erika M. Wolff, Yvonne C. Tsai, Wei Ye, Xianghong Zhou, and Peter A. Jones. The putative tumor suppressor microRNA-101 modulates the cancer epigenome by repressing the polycomb group protein EZH2. Cancer Res. 69, 2623-2629. (2009)
doi:10.1158/0008-5472.CAN-08-3114

89.Jian-Fu Chen, Elizabeth M Mandel, J Michael Thomson, Qiulian Wu, Thomas E Callis, Scott M Hammond, Frank L Conlon, and Da-Zhi Wang. The role of microRNA-1 and microRNA-133 in skeletal muscle proliferation and differentiation. Nat Genet. 38, 228-233. (2006)
doi:10.1038/ng1725

90.Lee Tuddenham, Guy Wheeler, Sofia Ntounia-Fousara, Jasmine Waters, Mohammad K. Hajihosseini, Ian Clark and Tamas Dalmay. The cartilage specific microRNA-140 targets histone deacetylase 4 in mouse cells. FEBS Lett. 580, 4214-4217. (2006)
doi:10.1016/j.febslet.2006.06.080

Abbreviations: PTM: Post Translational Modifications, miRNAs: micro RNAs, HP1: Heterochromatin Protein 1, DNMTs: DNA Methyl-Transferases, HDACs: Histone De-ACetylases, HMTs; Histone Methyl-Transferases, HATs: Histone Acetyl-Transferases, MECP2: Methyl CpG binding protein2, PRCs: Polycomb Repressive Complexes, MAGE: melanoma-associated antigen, MASPIN: SERine Protease INhibitors, IGF2: Insulin-like growth factor 2, MLH1: mutL homolog 1, BRCA1: Breast Cancer 1, RUNX3: runt-related protein, PML: promyelocytic leucemia, RAR: Retinoic Acid Receptor, PcG: Polycomb Group, EZH2: Enhancer of zeste homolog 2, RING1: ring finger protein 1, AML: Acute Monocytic Leucemia, MLL: Mixed Linear Leukaemia, CLL: chronic lymphocytic leucemia, HMGA2: the High-Mobility Group A2, PTEN: phosphatase and tensin, PDCD4: programmed cell death 4, TPM1: Tropomyosin, TIMP3: Metalloproteinase inhibitor 3, Bak1: BCL2-antagonist/killer 1, BCL2: B-cell lymphoma protein 2, NB: neuroblastoma, MB: medulloblastoma, TrkC: tyrosine kinase, KIT: c-kit, HER-2: Receptor Human Epidermal growth factor Receptor 2, ERBB2: erythroblastic leukemia viral oncogene homolog 2,

Key Words: Epigenetic, Chromatin, Histone PTMs, DNA-Mehtylation, miRNA, Cancer, Differentiation, Proliferation, Review

Send correspondence to: Patrizia Filetici, Istituto di Biologia e Patologia Molecolari, c/o Ist. Fisiologia Generale, Sapienza Universita di Roma, P.le A.Moro 5, 00185 Roma, Italy, Tel: 0039-0649912241, Fax: 0039-064440812, E-mail:patrizia.filetici@uniroma1.it