10.4 Epigenetics

Click to collapse Click to expand
Main | Save Edit | Discussion | History | Cube (0)

Epigenetic Mechanisms


Despite having nominally identical DNA sequences, clonally derived cells within a multicellular organism express diverse subsets of genes within their differentiated states. This non-genetic regulation is the basis of epigenetics; heritable features of the genome that can be influenced by the cell’s environment and are not directly coded within the DNA (1). Eukaryotic cells possess several epigenetic mechanisms responsible for a wide range of effects on gene expression. These epigenetic features include the chemical modification of the DNA through the methylation of cytosine bases (2), modification and remodeling of the overlying histone and other protein components of chromatin (3), and post-transcriptional gene regulation by small RNAs. Taken together, these mechanisms can be influenced by both cellular and environmental factors, from early development to continual modulation within the cell’s differentiated state.


As a disease of dysregulation, many of the aberrations leading to cancer are of epigenetic nature. Epigenetic marks leading to the silencing of tumor suppressor genes has become the predominant mechanisms, however, cases involving the upregulation of proto-oncogenes, with cancer causing abilities at when overexpressed, have been described. Deposition of these marks as well can be a direct function of the cell’s environment, dismissing the notion that mutagens are the only carcinogen.


In this section we will discuss the importance of promoter methylation, histone modifications and non-coding RNA, going into detail on their individual mechanisms and their possible roles in cancer development.


DNA methylation


An important epigenetic mark capable of regulating transcription, DNA Methylation is the chemical addition of a methyl group to cytosine residues within the DNA. In eukaryotic cells, cytosines targeted for methylation are often adjacent to guanosine nucleotides, the two bases constituting a CpG dinucleotide. In the mammalian genome, 60-90% of CpGs are methylated, with the highest density of this mark occurring on gene open reading frames (4). Among the predominately methylated, scattered CpGs within the genome are stretches of high CpG density, termed CpG islands (CGIs), which show a frequent absence of methylation (5). Averaging 1000bp, CGIs are have been shown to be localized at the 5’ regulatory regions of transcriptionally active genes, a correlation that has lead to the designation of methylated DNA has a repressive mark (4)(6).


DNA methylation patterning is thought to be a result of the high mutagenic properties of methylcytosine, which converts to thymine through spontaneous deamination. Therefore, an underrepresentation of CpG dinucleotides is present outside of CGIs, where they are often methylated and prone to mutation. In organisms such as invertabrates, where little or no DNA methylation takes CpGs occur at a much higher frequency throughout the genome. In affect the whole genome of these organisms is CGI like, showing the same increased CpG density, suggesting that the nucleotide patterning is a consequence of selective DNA methylation (4)(7).


Apart from the correlation with transcriptional activity, our understanding of functional significance of DNA methylation on gene expression remains incomplete. Studies suggest that unmethylated CGIs themselves act as gene promoters, and are able to recruit general transcription factors regardless of classic promoter motif content, such as the TATA box; in general, protein-binding sites within DNA have been shown to by more GC-rich than non-binding sequence. Further evidence from transient reporter gene assays has shown that these regions when associated with actively transcribed genes are enriched for a number general transcription factors (8)(9).


Recent work has established a direct link between CGIs and the transcriptionally active associated histone modification H3K4me3 (20). Studies show that Cfp1, a member of the histone methyltransferase complex Setd1, binds exclusively to unmethylated CpGs, depositing the H3K4me3 mark at these regions which can further promote transcrption through the recruitment of chromatin remodeling complexes, transcription factors and other histone modifiers (11).


Approximately 50% of CPIs are located within the promoter region of genes, while the remaining “orphan” CGIs are shared between intergenic and intragenic regions, and generally possess an increased methylation state (7). It has been shown that when demethylated, these orphan regions do possess transcription regulatory activity, potentially marking prior uncharacterized promoter regions of non-coding RNA molecules. Another functional example of orphan regions is CGI contained within intron 2 of Igf2r, a gene silenced on the parental chromosome through an imprinting mechanism. Here, the non-coding transcript Air is able to interact with the CGI leading to inactivation of Igf2r (12).


Studies into development showed that the methylation of CGI promoters controlling expression of early developmental genes effectively silenced them during later stages of development (13). Two theories exist to describe this silencing mechanism: a steric inhibition model, in which transcription factors are blocked from binding due to the methylated state of the promoter, and a repressor recruitment model, in which the high density of methylated is able to bind factors promoting gene silencing. Both models are not exclusive to the other, and evidence has been suggested for both (4).

Figure 10.4.1. Methylation of CpG islands in TSG promoters can suppress gene expression and lead to cancer. Released under the Creative Commons Attribution-ShareAlike 4.0 International license (CC BY-SA 4.0).


DNA methylation has been extensively researched in cancer cells, with many cancer specific methylation sites mapped within the human genome. At the basis of most described mechanisms is the aberrant methylation of CGI promoters functioning tin the regulation of tumor suppressor genes. What remains to be uncharacterized is methylation’s role in cancer initiation, and determining whether it is causative or a result of prior aberrations. Much work has been conducted into comparing methylation in cancerous and normal cells to observe whether it is consistent between the two. Differences have been observed in colorectal cancer cells, where steady methylation was documented over both CGIs present at the 5’ end of genes as well as at orphan CGIs, where methylation is not normally present (14).


Cancer associated methylation of CPI promoters includes the silencing of: p53 cell cycle inhibitor, p16 cyclin-dependent kinase inhibitor, MGMT tumor suppressor, APC cell cycle regulator and BRCA1 a DNA-repair gene.


As a separate mechanism of repression, CGI promoters can be silenced by polycomb group proteins (PcGs). Though not fully characterized it is known that two PcGs function together in silencing the mechanism: polycomb-repressive complex 2 possesses histone methylation capabilities and deposits the repressive HrK27me3 mark at CGIs. This mark recruits polycomb-repressive complex 1, which functionally inhibits transcription by an unknown mechanism involve H2A ubiquitination (25). The PcG silencing mechanism has been found to play a significant role during development, where CGI promoters in embryonic stem cells possess both H3K4me3 transcriptional activating marks and H3K27me3 deposited by PcGs. These cells are in a pluripotent state, due to the bivalent nature of the histone marks, which is lost during differentiation, where one of marks is removed either permanently activeating or silencing the gene (16). Remarkably, in some cancer cells, a similar mechanism have been described, providing pluripotency to varying degrees, favoring unrestrained proliferation of cancer cells (17).


Methylation of intrageneic CpGs, as well possibly lead to malignancies; methylcytosine’s tendency to spontaneously undergo deamination and convert to thymine mutates the base, the effect of which is variable, depending on where in the gene the CpG is located. UV absorption of cytosine is also altered in its methylated state, potentially leading to pyrimidine dimers, and a need for DNA damage repair (18).



Histone modification



Within eukaryotic cells, DNA is compacted into chromatin, consisting of the DNA itself, along with RNA and a number of associated proteins (3). Often likened to beads along a string, the basic unit of chromatin is the nucleosome, comprised of DNA wrapped around an octomer of histone proteins constructed from dimers of four different core histone types, H2A, H2B, H3 and H4. 147bp of DNA, comprising 1.6 turns, enwraps each histone octomer, with stretches of linker DNA with lengths of to 80 bp between each particle. Histone H1, its isoforms, and other non-histone proteins act within these linker regions to form chromatin’s secondary structures, further compacting the DNA from the beads along a strong model of nucleosomes arrangement (1)(3).


Each core histone protein is comprised of a globular region, comprising the bulk of the histone, and an n-terminal, unstructured tail region of about 20-30 amino acids long, which acts as a scaffold for many post translational modifications to the proteins (1). Covalent modifications to both the histone tail and globular regions include: acetylation, phosphorylation, methylation, and the addition of the small proteins ubiquitin and sumo. Together, these modifications affect the expression status of genes-whether they are actively transcribed or repressed, by influencing the dynamic structure of chromatin to a more open euchromatin, or condensed heterochromatin form respectively. Further, the influence of these histone marks extends to a host of other genomic processes, contingent on chromatin conformation, such as DNA replication and damage repair.


Characteristic patterning of histone modifications is observed along stretches of DNA, correlating with their recognized functions in modulating chromatin state. Trimethylation of the lysine 4 residue of the histone H3 N-terminal tail region (H3K4me3) (3). for example is found specifically within the promoter and 5’ transcribed regions of actively transcribed genes, while the repressive mark H3K27me3 (3). is associated with transcriptional repression and is often mapped to genes not transcriptionally active. It is no surprise, therefore, that these characteristic patterns of histone modification have been found greatly altered within cancer cells, where transcriptional regulation is aberrant. 


Here, detail will be provided for four major histone modifications: acetylation, methylation, phosphorylation and ubiquitination, describing the different roles prescribed to each mark in genetic physiology, with a focus on the role of gene expression. These four marks have been studied extensively within both cancer and basic research fields, and much has been elucidated concerning their functions both within cancerous and non-cancerous cells.


Histone Acetylation


The addition of an acetyl group to the N-terminal tail or histone globular region changes the overall charge of the protein from positive to neutral (19). In general, the removal of histone charge influences nucleosome formation and chromatin compaction, which is partially dependent on the affinity between the opposing charges of DNA and histones. Interactions between the histone octomer and DNA are lessened in the process, resulting in relaxation of chromatin state, and increased access to the DNA strand by transcription factors. Through this mechanism, acetylation at specific histone residues appears to be less important than the overall histone acetylation state of a gene on the whole with no single lysine reside being more or less critical than the next. In a second mechanism of transcriptional activation, acetylated histones can themselves be the target of transcription factor binding through the bromodomain protein domain, which enables specific binding to the modified histone. Bromodomains act as acetyl-lysine recognition modules, directing enzymes that contain them to particular chromosomal sites (3)(19). The targeting of binding factors has implicated the histone acetylation in a number of roles, apart from transcriptional activation, including DNA damage repair and replication. As a modulator of gene expression, histone acetylation has been shown to play a number of roles in establishing the proper transcriptome of a cell during the four stages of the cell cycle. It is therefore crucial for these marks to be properly added and removed, whereas mutations affected their patterning can both prevent and induce cell cycle progression, establishing malignant cell types (1)(3)(19). 


Histone acetylation is catalyzed by histone acetyltransferases (HATs), whereas removal of the modification is performed by histone deacetylases (HDACs) (20). Several HATs and HDACs exist within the eukaryotic genome, conserved from yeast to humans, with lysine substrate specificity contingent on the enzyme, the organismal homolog and the complex by which it acts through. The conserved HAT, Gcn5, demonstrates HAT substrate promiscuity between organisms, and the complexity in determining how histone acetylation in targeted to specific chromatin regions (21)(22). Functioning in vitro, within the complex SAGA (Spt-Ada-Gcn5 acetyltransferase), Gcn5 regularly acetylates histone H3 at lysine residues K9, K14, K18, and K23. However, in vivo studies within Drosophila larva have shown Gcn5 activity at H4K5 and H4K12; in chicken cells, Gcn5 deletion mutants loose acetylation at H2BK16; and disruption of the enzyme in mice leads to decreased acetylation in all aforementioned residues, except H3K14. In addition to functioning within SAGA, human Gcn5 is also found within the HAT complex ATAC, which shares some target lysine residues with SAGA but uniquely acetylates others, and is targeted to chromatin by a different manner (22)(23).


Histone Acetylation and Cancer


Although functional redundancy is apparent between enzymes, specific roles have been prescribed for a number of HATs and HDACs at all levels of organism complexity. It is through both the primary association of histone acetylation with transcriptional activity, and its more recently prescribed secondary roles, that researchers have given a functional basis to the correlation between mutations and altered expression of HAT and HDAC enzymes and cancer aggression.


TRRAP, a component of the SAGA complex, acts as a necessary co-factor for the transcriptional regulatory function of the proto-oncogenic transcription factor c-myc. Through the interaction of these two factors, c-myc can recruit GCN5, the catalytic subunit of SAGA, to gene regulatory sites to induce transcriptional activity. The tumor suppressor BRCA1, which functions in DNA repair and cell cycle progression, binds GCN5 through TRRAP in a similar manner. Mutations affecting this interaction have been shown to predispose women to early onset of breast cancer, suggesting that the HAT activity of GCN5 is necessary for BRCA1-mediated gene regulation and DNA repair (25).


In one of the first published studies to describe the impact of histone acetylation on cellular transformation, Brehm et al. showed that the histone deacetylase HDAC1 associates with tumor suppressor, retinoblastoma protein (pRb), to mediate E2F-bound promoter repression and silencing of genes needed to progress into the cell cylce S phase. When pRb binds E2F, HDAC1 is present within the complex, which removes expression-activating histone acetylation marks from genes bound by E2F. pRb incapable of this function are not able to transport HDAC1 to the E2F regulated genes, causing aberrant expression of S phase related factors, and cell transformation (26).



Histone Methylation


Unlike histone acetylation, which is associated only with transcriptional activation, methylation of the lysine residues of histone tails and globular regions can be associated with both gene activation and repression, depending on the specific histone residue modified. Furthermore, histone residues can be mono-, di- or tri- methylated, each mark having a potentially different function associated (3)(27). The addition of methyl groups does not change the overall histone charge, and it is unlikely that the modification has any direct consequences on nucleosome stability and chromatin folding. Instead, methylated histones function solely through the recruitment of specific chromatin binding proteins, which interact through one of many binding domains specific to methylated histones (3)(27) These binding domains include: PHD fingers, Tudor domains, chromodomains, and WD-40 domains, each capable of binding a host of unique methylated lysine marks. Many functions have been prescribed to these factors apart from transcriptional activation and repression, including chromatin remodeling, DNA damage repair, and crosstalk with other epigenetic modifiers (3)(28).


Although a complete mechanism for function of many histone methylation marks has yet to be completely mapped, strong correlations exist between specific marks and actively expressed, or unexpressed genes. For example, Trimethylation of histone H3 at lysine 4 (H3K4me3) is known to be associated with transcriptional activation of the underlying gene, while trimethylation of histone H3 at lysine 27 appears to be associated with the gene’s transcriptional repression. On top of these correlations, the specific marks are targeted and function at different regions of the gene (3)(28). In regards to H3K4 methylation, the three levels of methyl group content are patterned across the gene landscape from H3K4me3 at the gene 5’ end to H3K4me1 at the 3’. Although other mechanisms exist, patterning of these marks often arises through the association and disassociation of different histone methyltransferase (HMT) enzymes to RNA polymerase as it travels along an actively transcribed gene. During transcription the carboxy terminal domain (CTD) of RNA polymerase II, is differentially phosphorylated by associating factors of the transcriptional machinery. These different CTD phosphorylation states promote the binding of different histone modifiers, including HMTs, as the gene is transcribed, creating characteristic patterns of modifications along the open reading frame and promoter elements (3)(28). Although not fully characterized as well, it appears that this patterning plays a role in how histone modification binding proteins interact with the chromatin landscape, ultimately affecting the modification’s downstream function. 


Histone Methylation and Cancer


Due to the variety of functions prescribed to histone methylation, many diseases, including a number of cancers, have been shown to be associated with the mark. Most of these diseases appear to stem from dysregulation of the characteristic histone methylation gene patterning of healthy cells. Using chromatin immunoprecipitation methods, it has been shown that within cancer cells, histone methylation marks that are correlated with different transcriptional activities are often absent, present in decreased or increased amounts, or are targeted to completely different gene subsets. The source of such irregular modification patterning within these cells is thought to be due to a number of mechanisms capable of a high level of crosstalk, however our understanding remains incomplete (29)(30).


MLL1 (ALL-1) is a methyltransferase of H3K4, which resides at a locus frequently rearranged in human lymphoid and myeloid leukemias. Translocations at this site were some of the first genetic abnormalities to show the link between histone methylation and cancer progression. Over 50 different translocations of MLL1 have been described, each showing different cancer topography and severity in harboring patients. The most common translocations are thoughs that remove the catalytic domain of MLL1 itself, rendering the protein unable of carrying out its methyltransferase activity. Attenuated H3K4 methylation results in transcriptional dysregulation of genes regulated by MLL1. Most noticeably these expression defects include those of the Hox gene family, which frequently observed in human cancers to have altered expression patterns (30).


Epigenetic Writers EZH2 and MLL2 in Cancer


The histone tri-methylase EZH2 has been implicated in various forms of cancer while displaying both tumor suppressive and oncogenic activity (31). EZH2 overexpression is involved in silencing of anti-proliferative genes in diffuse large B-cell lymphoma (DLBCL) (32). Alternatively, in other DLBCL and follicular lymphoma (FL) cases, somatic mutations and deletions of EZH2 have been observed to decrease the protein’s activity (33). The fusion oncoprotein EWS-FLI1 transcriptionally activates EZH2 expression in Ewing’s Sarcoma (34). Moreover, in numerous types of cancer EZH2 has also been observed to be overexpressed, including prostate cancers, melanoma, and bladder cancer (31).


As the main catalytic subunit of the Polycomb Repressor Complex 2 (PRC2), EZH2 mediates transcriptional repression via its methyltransferase activity. EZH2 tri-methylates histone tails at histone3 lysine27 (H3K27) residues, resulting chromatin condensation and epigenetic silencing of target gene transcription (31). Genes targeted by EZH2 have roles in regulation of cell cycle checkpoints, DNA damage repair, cell fate and differentiation, offset of senescence, and apoptosis (31). Therefore, dysregulation of EZH2 activity can act as a driver of cancer transformation and metastasis.


EZH2 activity represses the activity of several tumor suppressor genes as well as genes implicated in tumourigenesis. The cell cycle regulator INK-ARF is supressed by EZH2 activity, enhancing cell cycle progression and preventing senescence in metastasis (34). EZH2 has been shown to induce cancer progression by targeting E-cadherin for repression, promoting epithelial-mesenchymal transition (34). EZH2 overexpression has also been observed to enhance metastasis by inducing both NFκB and Ras pathways; a result of EZH2’s transcriptional silencing of the RasGAP protein, Disabled Homolog2-Interacting Protein (DAB2IP) (34).


Alternatively, the epigenetic writer MLL2 induces the transcription of its target genes. MLL2 lays tri-methyl marks at histone3 lysine4 (H3K4) prompting the relaxation of chromatin structure and expression of associated genes (35). Genes targeted by MLL2 are implicated in cell migration, growth adhesion, and transcriptional regulation (36). Knockout of MLL2 has been observed to reduce growth and increase apoptosis in mice and cell lines (37).


Like EZH2, MLL2 has both tumor suppressor and oncogenic functions. Genome sequencing has revealed MLL2 nonsense and loss-of-function mutations are common in FL and DLBCL (38). However, amplifications and translocations have been frequently observed in leukemia as well as solid malignant tumours (36). Loss-of-function mutations in MLL2 are also heavily associated with the congenital disorder Kabuki syndrome, characterized by several malformations and intellectual disability (39).




(1) S.L. Berger, T. Kouzarides, R. Shiekhattar, A. Shilatifard. 2009. An operational definition of epigenetics genes Dev., 23. 781–783

(2) R.S. Illingworth, A.P. Bird. 2009. CpG islands—“a rough guide”. FEBS Lett., 583. 1713–1720

(3) S.L. Berger. The complex language of chromatin regulation during transcription. 2007. Nature, 447. 407–412

(4) Robertson, K. DNA methylation and human disease. Nature Reviews Genetics 6, 597–610

(5) Bird A, Taggart M, Frommer M, Miller OJ, Macleod D. 1985. A fraction of the mouse genome that is derived from islands of nonmethylated, CpG-rich DNA. Cell 40: 91–99.

(6) Suzuki MM, Kerr AR, De Sousa D, Bird A. 2007. CpG methylation is targeted to transcription units in an invertebrate genome. Genome Res 17: 625–631.

(7) Tucker KL (June 2001). "Methylated cytosine and the brain: a new base for neuroscience". Neuron 30 (3): 649–652

(8) Choi JK. 2010. Contrasting chromatin organization of CpG islands and exons in the human genome. Genome Biol 11: R70. 

(9) Carninci P, Sandelin A, Lenhard B, Katayama S, Shimokawa K, Ponjavic J, Semple CA, Taylor MS, Engstrom PG, Frith MC, et al. 2006. Genome-wide analysis of mammalian promoter architecture and evolution

(10) A chromatin landmark and transcription initiation at most promoters in human cells. Cell 130: 77–88

(11)CpG-binding protein (CXXC finger protein 1) is a component of the mammalian Set1 histone H3-Lys4 methyltransferase complex, the analogue of the yeast Set1/COMPASS complex. J Biol Chem 280: 41725–41731.

(12) Sleutels F, Zwart R, Barlow DP. 2002. The non-coding Air RNA is required for silencing autosomal imprinted genes. Nature 415: 810–813. 

(13) Mohn F, Weber M, Rebhan M, Roloff TC, Richter J, Stadler MB, Bibel M, Schubeler D. 2008. Lineage-specific polycomb targets and de novo DNA methylation define restriction and potential of neuronal progenitors. Mol Cell 30: 755–766.

(14) Illingworth RS, Gruenewald-Schneider U, Webb S, Kerr ARW, James KD, Turner DJ, Smith C, Harrison DJ, Andrews R, Bird AP. 2010. Orphan CpG islands identify numerous conserved promoters in the mammalian genome. PLoS Genet 6: e1001134. doi: 10.1371/journal.pgen.1001134. 

(15) Stock JK, Giadrossi S, Casanova M, Brookes E, Vidal M, Koseki H, Brockdorff N, Fisher AG, Pombo A. 2007. Ring1-mediated ubiquitination of H2A restrains poised RNA polymerase II at bivalent genes in mouse ES cells. Nat Cell Biol 9: 1428–1435.

(16) Mohn F, Weber M, Rebhan M, Roloff TC, Richter J, Stadler MB, Bibel M, Schubeler D. 2008. Lineage-specific polycomb targets and de novo DNA methylation define restriction and potential of neuronal progenitors

(17) Ohm JE, McGarvey KM, Yu X, Cheng L, Schuebel KE, Cope L, Mohammad HP, Chen W, Daniel VC, Yu W, et al. 2007. A stem cell-like chromatin pattern may predispose tumor suppressor genes to DNA hypermethylation and heritable silencing. Nat Genet 39: 237–242.

(18) Jones PA, Baylin SB (2002). "The fundamental role of epigenetic events in cancer.". Nat Rev Genet 3 (6): 415–28.

(19) Phillips DM. 1963. The presence of acetyl groups in histones. Biochem J. 87:258-63.

(20) Brownell JE, Allis CD. 1995. An activity gel assay detects a single, catalytically active histone acetyltransferase subunit in Tetrahymena macronuclei. Proc Natl Acad Sci U S A. 92:6364-8.

(21)  Brownell JE, Zhou J, Ranalli T, et al. 1996. Tetrahymena histone acetyltransferase A: a homolog to yeast Gcn5p linking histone acetylation to gene activation. Cell. 84:843-51.

(22) Lee KK, Workman JL. 2007. Histone acetyltransferase complexes: one size doesn’t fit all. Nat Rev Mol Cell Biol.8:284-95.

(23) Hildmann C, Riester D, Schwienhorst A. 2007. Histone deacetylases: an important class of cellular regulators with a variety of functions. Appl Microbiol Biotechnol.75:487-97.

(24) Dhalluin C, Carlson JE, Zeng L, et al. 1999. Structure and ligand of a histone acetyltransferase bromodomain. Nature.399:491-6

(25) McMahon S, Wood M, and Cole MD. 2000. The Essential Cofactor TRRAP Recruits the Histone Acetyltransferase hGCN5 to c-Myc. Mol Cell Biol. 20(2): 556–562.

(26) Brehm A, Miska EA, McCance DJ, Reid JL, Bannister AJ and Kouzarides T. 1998. Retinoblastoma protein recruits histone deacetylase to repress transcription. Nature 391, 597-601.

(27) Izzo A, Schneider R. 2010. Chatting histone modifications in mammals. Brief Funct Genomics. 9:429-43.

(28) Taverna SD, Li H, Ruthenburg AJ, Allis CD, Patel DJ. 2007. How chromatin-binding modules interpret histone modifications: lessons from professional pocket pickers. Nat Struct Mol Biol. 14:1025-40.

(29) Bonasio R, Tu S, Reinberg D. 2010. Molecular signals of epigenetic states. Science. 330:612-6.

(30) Canaani E. Nakamura T. Rozovskaia S. Smith T. Mori T. Croce C. and Mazo A. 2004 ALL-1/MLL1, a homologue of Drosophila TRITHORAX, modifies chromatin and is directly involved in infant acute leukaemia. Br J Cancer. 90(4): 756–760.

 (31) M. Sauvageau, G. Sauvageau. 2010. Polycomb group proteins: multi-faceted regulators of somatic stem cells and cancer. Cell Stem Cell 7(3):299-313.

(32) I. Velichutina, R. Shaknovich, H. Geng, N.A. Johnson, R.D. Gascoyne, A.M. Melnick, O. Elemento. 2010. EZH2-mediated epigenetic silencing in germinal center B cells contributes to proliferation and lymphomagenesis. Blood 116(24):5247-55.

(33) R.D. Morin, N.A. Johnson, T.M. Severson, A.J. Mungall, J. An, R. Goya, J.E. Paul, B.W. Woolcock, F. Kuchenbauer, D. Yap, R.K. Humphries, O.L. Griffith, S. Sha, H. Zhu, M. Kimbara, P. Shashkin, J.F. Charlot, M. Tcherpakov, R. Corbett, A. Tam, R. Varhol, D. Smailus, M. Moksa, Y. Zhao, A. Delaney, H. Qian, I. Birol, J. Schein, R. Moore, R. Holt, D.E. Horsman, J.M. Connors, S. Jones, S. Aparicio, M. Hirst, R.D. Gascoyne, M.A. Marra. 2010. Somatic mutations altering EZH2 (Tyr641) in follicular and diffuse large B-cell lymphomas of germinal-center origin. Nat Genet 42(2):181-5.

(34) C.J. Chang, M.C. Hung. 2012. The role of EZH2 in tumour progression. Br J Cancer 106(2):243-7.

(35) T. Chen, S.Y. Dent. 2014. Chromatin modifiers and remodellers: regulators of cellular differentiation. Nat Rev Genet 15(2):93-106.

(36) D.G. Huntsman, S.F. Chin, M. Muleris, S.J. Batley, V.P. Collins, L.M. Weidmann, S. Aparicio, C. Caldas. 1999. MLL2, the second human homolog of the Drosophila trithorax gene, maps to 19q13.1 and is amplified in solid tumor cell lines. Oncogene 18(56):7975-84.

(37) S. Cambacher, M. Hahn, G. Schotta. 2010. Epigenetic regulation of development by histone lysine methylation. Heredity 105(1):24-37.

(38) R.D. Morin, M. Mendez-Lago, A.J. Mungall, R. Goya, K.L Mungall, R.D. Corbett, N.A. Johnson, T.M. Severson, R. Chiu, M. Field, S. Jackman, M. Krzywinski, D.W. Scott, D.L. Trinh. J. Tamura-Wells, S. Li, M.R. Firme, S. Rogic, M. Griffith, S. Chan, O. Yakovenko, I.M. Meyer. E.Y. Zhao, D. Smailus, M. Moksa, S. Chittarajan, L. Rimsza, A. Brooks-Wilson, J.J. Spinelli, S. Ben-Neriah, B. Meissner, B. Woolcock, M. Boyle, H. McDonald. A. Tam, Y. Zhao, A. Delaney, T. Zeng, K. Tse, Y. Butterfield, I. Birol, R. Holt, J. Schein, D.E. Horsman, R. Moore, S.J. Jones, J.M. Connors, M. Hirst, R.D. Gascoyne, M.A. Marra. 2011. Frequent mutation of histone-modifying genes in non-Hodgkin lymphoma. Nature 476(7360):298-303.

(39) Y. Bokinni. 2012. Kabuki syndrome revisited. J Hum Genet 57(4):223-7.