Perspective Epigenetics: the language of the cell? · 2014-10-03 · may functionally acquire such a unique language characteristic. In summary, ... processes within the cell and

73ISSN 1750–1911Epigenomics (2014) 6(1), 73–88

Epigenetics is one of the most rapidly developing fields of biological research. Breakthroughs in several technologies have enabled the possibility of genome-wide epigenetic research, for example the mapping of human genome-wide DNA methylation. In addition, with the development of various high-throughput and high-resolution sequencing technologies, a large number of functional noncoding RNAs have been identified. Massive studies indicated that these functional ncRNA also play an important role in epigenetics. In this review, we gain inspiration from the recent proposal of the ceRNAs hypothesis. This hypothesis proposes that miRNAs act as a language of communication. Accordingly, we further deduce that all of epigenetics may functionally acquire such a unique language characteristic. In summary, various epigenetic markers may not only participate in regulating cellular processes, but they may also act as the intracellular ‘language’ of communication and are involved in extensive information exchanges within cell.

Keywords: cell language • cellular processes • ceRNA • communication • cross-talk • epigenetic markers • epigenetics • immune system • modifications • ncRNA

Many genetic phenomena may be reason-ably explained by Mendelian laws. Since DNA is identified as the genetic material of an organism, this discovery further provided the molecular basis for this type of classical genetics. However, it has long been recog-nized that there are other genetic phenomena that exhibit non-Mendelian inheritance pat-terns. For example, the selective inactivation of the X chromosome during early embryo development in female mammals [1], para-mutations (including mutations of the corn R gene and morphology of the Arabidopsis flowers) [2], the piebald phenomenon of the Drosophila position effect [3], and the recent discovery of the dosage compensation effect [4], chromosome imprinting [5] and prion-induced diseases [6], among others. Conrad Waddington introduced the term ‘epigenetic landscape’, which meant the transformation of genetic information into observable char-acteristics and the molecular basis of its phe-notype [7]. In many contexts, the epigenetic

patterns of gene expression and their associ-ated phenotypes are maintained after mitosis or meiosis. Thus, various epigenetic mark-ers constitute the basis of epigenetics, they are involved in regulating different kinds of processes within the cell and affect cellular activity in a potentially heritable way, which finally achieving cellular answer to changes of extracellular or intracellular environment and the development of multicellular organ-ism in a suitable way [8].

Epigenetics exhibit two distinct features: two phenotypes (closed and open); and ability to self-reinforce or form positive and negative feedback pathways [9]. In terms of two phenotypes of epigenetics, it means that epigenetics can mediate the occurrence of a single event. For example, high methylation of DNA at a promoter site frequently induces gene silencing, which further induces the formation of condensed chromatin. In this condition, epigenetics displays the closed phenotype, inhibiting the occurrence of cel-

Epigenetics: the language of the cell?

Biao Huang1, Cizhong Jiang2 & Rongxin Zhang*,1

1Research Center of Basic Medical

Science, Department of Immunology,

Basic Medical College, Tianjin Key

Laboratory of Cellular & Molecular

Immunology, Key Laboratory of Immune

Microenvironments & Diseases of

Educational Ministry of China, Tianjin

Medical University, Tianjin 300070,

China2Department of Bioinformatics, Shanghai

Key Laboratory of Signaling & Disease

Research, The School of Life Sciences &

Technology, Tongji University, Shanghai,

China

*Author for correspondence:

Tel.: +86 22 83336563

Fax: +86 22 83336563

[email protected]

part of

Perspective

10.2217/EPI.13.72 © 2014 Future Medicine Ltd

Perspective

74 Epigenomics (2014) 6(1)

lular processes. Comparatively, the open phenotype of epigenetics usually mediates the upregulation of cel-lular processes. For example, histone acetylation fre-quently leads to the formation of unwound chromatin, which increases the accessibility for various transcrip-tion factors. Moreover, after the disappearance of the initial stimulus signal, the established epigenetic pat-terns are still able to achieve self-sustainment through various cis- and trans-regulatory mechanisms. Via self-enhancement, the epigenetic state in trans is predomi-nantly maintained through the formation of positive feedback loop pathways and complex transcription fac-tor networks, which result in the establishment of sta-ble transcriptional states of self-renewal. For example, development-related transcription factors and ncRNAs (particularly regulatory sRNAs and lncRNAs) [10]. However, other trans-epigenetic markers can self-renew via enzyme catalysis. For example, self-catalysis of prion [11] and special ncRNAs may provide sub-strates for unique enzymes (e.g., the RNA-dependent RNA polymerase (RdRP) in plants and worms [12]). After each cell division, the trans-epigenetic signal that is inherited from the mother cell mediates signal self-proliferation in the daughter cells, achieving the main-tenance of trans-epigenetic markers at a specific level within the daughter cells. In comparison, cis-epigenetic markers physically interact with chromatin and some-how self-renew coupled with chromosome replication, such as DNA methylation, various histone modifica-tions and nucleosome positioning, whereas some occur independently of replication such as histone variant replacement. When cell division occurs with the rep-lication of genomic DNA, the cis-epigenetic markers are also somehow duplicated so that the cell-specific cis-epigenetic patterns are stably maintained even though it is still unknown how histone modifica-tions, histone variants and nucleosome positioning are inherited [13–16].

At the molecular level, the trans-epigenetic mecha-nisms, which are mainly based in the cytoplasm, enable the regulation of various cytoplasmic processes by various trans-acting regulatory ncRNAs [17] and other functional proteins [18,19] that are involved in the reg-ulation of gene expression at the post-transcriptional level or the maintenance of cell type-specific charac-teristics. However, cis-epigenetic mechanisms mainly based in chromatin include DNA cytosine methylation and 5-hydroxymethylcytosine [20,21], covalent modifi-cations in various histones (H1, H2A, H2B, H3 and H4) and histone variants, replacement of histone vari-ants [22], nucleosome positioning and various cis-acting regulatory ncRNAs that mediate chromosome activa-tion or silencing, among others. These cis-epigenetic mechanisms are involved in the regulation of various

events that occur at the chromatin level, including dis-tinct phases of gene transcription, pre-mRNA mature splicing and alternative splicing, mRNA monitoring (e.g., control of nuclear export of the mature tran-script), DNA damage repair and DNA replication, among others.

However, various intracellular processes, including those processes that are closely linked with various epigenetic markers, are not interdependent within the cell, especially during the development of an organism (e.g., the inter-relationship of cell proliferation, differ-entiation and apoptosis). For example, when the cell is undergoing cell division, the expression of cell divi-sion-related genes is greatly enhanced. In addition, the expressions of these genes that promote apoptosis are inhibited. Another typical example occurs during the development of the immune system where under Th1 cell polarization conditions, undifferentiated CD4+ T cells are directionally differentiated into Th1 cells. In such a process, the expression of some Th1 cell-specific genes, including IFN-γ, Stat4 and T-bet, are upregulated; however, the expression of Th2 and Th17 cell-specific genes such as Gata3 and Il17a-Il17f are further inhibited, resulting in the stable establishment of various Th1 cell-specific biological characteristics [23]. In addition, it has been shown that site specific-epigenetic markers are involved in the fine-turning of gene expression, thus raising the question: do these epigenetic markers participate in synergistic expression among related genes during the development? Do all cis- or trans-epigenetic mechanisms also mediate this type of synergistic response among various cellular processes? If so, by which means do they achieve this type of synergistic response?

Recently, the ceRNA hypothesis revealed an inter-esting point: miRNAs not only mediate the occur-rence of a single event (e.g., miRNAs regulate mRNA degradation or inhibition of translation) but also act as the ‘language’ to mediate the mutual communica-tion among mRNAs that are regulated by the same unique miRNA, thus achieving integral regulation and improving regulatory efficiency [24–26]. This type of language property may be shared with all ncRNAs, more than miRNAs.

In this review, on the basis of two distinct epigen-etic features, we propose that noncoding RNAs, spe-cifically regulatory ncRNAs, also function as a type of specific epigenetic marker. In addition, we gained inspiration from the recent proposal and confirma-tion of the ceRNA hypothesis, we propose that the ncRNAs’ ‘language’ ability is shared with other epi-genetic markers. Thus, we put forward the core idea of this review: that epigenetics is the language of the cell. Perhaps it is this language ability of epigenetics

future science group

Perspective Huang, Jiang & Zhang

www.futuremedicine.com 75

that mediates the synergistic responses among various cellular processes.

Epigenetics, in terms of DNA methylation (special-site cytosine of DNA is methylated into 5-methylcyto-sine, which recruits functional complexes containing DNA-methylation domains, further achieving tran-scriptional inhibition), histone modifications (histones can be phosphorylated, methylated, acetylated or ubiq-uitinated, these histone modifications can somehow promote and suppress gene expression) and histone variants (histone variants are very similar to typical histone with slight difference, they are involved in establishing the special domains of chromatin, medi-ating gene expression regulation), and nucleosome positioning, has been systematically described in many studies [27–31]. In this review, we will mainly focus on the role of ncRNAs in epigenetics and. accordingly, further clarify the core idea that epigenetics is the language of the cell.

ncRNAs: an important part of epigeneticsLower organisms, such as Caenorhabditis elegans, have a similar and comparable number of protein-coding genes to humans; however, the human genome is 30 times the size of the C. elegans genome. Thus, the developmental and physiological complexity of the human is hardly attributed to a small number of protein-coding genes in the genome. This also implies that non-coding sequences, which account for the most of the genome, play a vital role in directing the formation of higher biologically complex organisms. With the development of deep sequencing technology of the human genome, we now know that there are only approximately 20,000 protein-coding genes in the human genome, which occupy <2% of the sequences in the entire human genome. Perhaps alternative splic-ing of protein-coding genes’ pre-mRNAs and various post-translational covalent modifications of proteins greatly expand the information content of the protein-coding genes and improve diversity and functional-ity of the proteome. This may also partly explain the increased complexity observed in higher organisms. However, high-throughput sequencing technology has revealed that most of the DNA sequences in the mam-malian genome may be transcribed (in fact, >90% of the human genome may be transcribed). The major-ity of these transcripts comprise various regulatory sRNAs (sRNAs can be roughly divided into three classes: miRNAs, endo-siRNAs and piRNAs, which are ∼20–30-nucleotide small molecules and are closely coupled with the Ago protein family members, guid-ing Ago to special target sites, then typically leading to downregulation of the target gene expression) that are involved in the regulation of gene expression and

tens of thousands of lncRNAs (lncRNAs are arbitrarily identified as the >200 nucleotide ncRNAs molecules with no or little protein-coding potential; they can be folded into many functional secondary structures, through which these lncRNAs may exert their bio-logical functions), which do not or have little func-tion in protein coding. Instead, lncRNAs participate in various intracellular processes. Previous studies have shown that the transcriptional level of the noncoding sequences is at least four-times higher compared with coding sequences. In recent years, a large number of biologically functional ncRNAs have been success-fully identified [32], particularly lncRNAs that display functional diversity [33].

Importantly, many ncRNAs also play an important role in epigenetics. They are not only involved in the establishment of the epigenetic patterns of various epigenetic markers (e.g., special ncRNAs mediate the establishment of active or inhibitory epigenetic mark-ers at unique loci), but these ncRNAs are also a type of important epigenetic marker. These ncRNAs (particu-larly various regulatory ncRNAs) mediate the occur-rence of various intracellular processes. For example, lncRNA-HOTAIR mediates the establishment of inhibitory H3K27 me3 at the HOXD locus, thereby silencing expression [34]. Thus, some ncRNAs possess the switch ability to regulate an event on or off, such as activation or silencing of gene expression, stabili-zation or degradation of mRNA. Furthermore, these ncRNAs may also perform self-reinforcement through different mechanisms, forming positive and nega-tive feedback pathways. For example, some ncRNAs achieve self-renewal by enzyme catalysis (e.g., RdRPs) and piRNAs by the ping-pong pathway [35]; however, a number of ncRNAs may achieve self-reinforcement at the transcriptional level (e.g., they are involved in establishing stable epigenetic patterns at one’s own genetic loci or other loci whose expression products promote the stable expression of specific ncRNAs). Thus, these ncRNAs exhibit two classical epigenetic features because of their extensive transcription and diverse functions. Undoubtedly, these ncRNAs play an important role in epigenetics. A variety of cell type-specific regulatory ncRNAs (specifically those that involve the establishment of cell-type-specific epigen-etic patterns) are passed to daughter cells through cell division. These inherited ncRNAs regulate different types of intracellular processes by cis- or trans-acting modes, which ultimately stably establishes and main-tains various cell type-specific biological characteris-tics. It was recently reported that paramutations are closely linked with sRNAs [36–39].

According to the ceRNAs hypothesis, we propose that this type of language characteristic may be shared


Epigenetics: the language of the cell? Perspective

76 Epigenomics (2014) 6(1) future science group


Figure 1. Schematic diagram showing that the ncRNAs ‘brain’ operates various cellular processes on the base of the limited genome. All ncRNAs in the cell are comprised of a huge ‘ncRNA library’; it is not only involved in regulating various cellular processes, including cell proliferation (symbolized by the printer) [43], differentiation (the car loaded with O2 symbolizes an erythrocyte, the hoover symbolizes for various immune cells), apoptosis, mature splicing of pre-RNA, protein translation and genome integrity maintenance, among others; in addition, it mediates the extensive crosstalk of various processes, enabling the cellular whole response. These ncRNAs consist of an intangible cellular brain center, on the basis of the limited genome and genome’s products, the cellular brain directs the cell to make the optimal cellular reply in response to various extracellular and intracellular stimuli (the same type of ncRNAs are shown in different colors to display the diversity). B: B lymphocytes; FA: Fatty acid; HB: Hemoglobin; Mϕ: Macrophage; NA: Nucleic acid; NK: Natural killer cell; PCD: Programmed cell death; Pro: Protein; T: T lymphocytes.

G

A U C G

H S V K T Q

AAAm7G

aa aa

Pre-RNA splicing

Protein translation

SNK

C

H

Genome

Genome integrity maintenance

Cell proliferation

FA

CH2O Pro

NA

ApoptosisPCD

Bcl caspase

Erythrocyte

siRNApiRNA

piRNA

diRNA

lncRNA lncRNA

lncRNA

miRNA

miRNA

lncRNAmiRNA

lncRNAmiRNA

lncRNA

lncRNA

miRNAmiRNAmiRNA

miRNA

lncRNAmiRNA

lncRNAmiRNA

lncRNAmiRNA

endo-siRNA

endo-siRNA

snoRNA

rRNAtRNA

rRNAtRNA

snRNA

miRNA

lncRNA

lncRNA

miRNA

miRNAN

KNK

T

NK Mφ

Mφ

B

Immunocyte

TA

TMTs

Th

B

CTL

O2O2O2

O2O2O2O2

O2

O2 O2

HB

Cell membrane sensors detecting the change of extracellular environment, then passing the information into the ncRNA ‘brain’

Intracellular sensor detecting the change of intracellular environment, then passing the information into the ncRNA ‘brain’

ncRNAs


with various ncRNAs, more than miRNAs. By act-ing as the ‘language’ for communication, ncRNAs not only mediate extensive communication of genome-wide protein-coding information, but they may also connect with various intracellular processes, enabling the synergistic response at the scale of the whole cell, thus adapting to different types of varying intracel-lular and extracellular environments. For example, a language that consisted of housekeeping ncRNAs (e.g., tRNAs, rRNAs, snRNAs and snoRNAs, among oth-ers) may participate in the regulation of the maturation of pre-mRNAs and translation of mature mRNAs. A language that consists of miRNAs and lncRNAs may mediate the well-organized expression of the whole genome sequence. A language that consists of piRNAs [40], diRNAs [41] and endo-siRNAs [42] may participate in the maintenance of genomic integrity. This is analo-gous to a human-like brain center, which consists of various ncRNAs. This cellular ‘brain’ directs the cell to generate optimal responses to various intracellular and extracellular stimuli on the basis of the limited genome and genome products (Figure 1).

Extensive crosstalk among the various epigenetic markersDifferent types of cis- and trans-epigenetic markers do not work alone, regardless of the epigenetic markers in the cytoplasm or nucleus. There is extensive crosstalk among the intra- and inter-epigenetic markers, which form an ‘epigenetic network’ throughout the entire cell. The results that we observed are a mutually syn-ergistic additive effect of extensive crosstalk. Impor-tantly, the final result of crosstalk often forms numer-ous positive and negative feedback loops, which result in the formation of an ‘on’ or ‘off ’ state of numerous cellular processes, or may accelerate or slow down some processes. Various ncRNAs are also closely associated with other epigenetic markers, which form extensive crosstalk.

Crosstalk between noncoding RNAs & other epigenetic markersThere is a common relationship of reciprocal induc-tion between ncRNAs and DNA methylation. For example, RNA-directed DNA methylation in plants is achieved via two steps: the generation of double-stranded RNA, which is further processed and pack-aged into a RITS complex; and an interaction between RITS and DNA methyltransferases (DNMTs), which target the DNMTs to specific sites (e.g., gene promot-ers or repetitive sequences) to mediate de novo DNA methylation [44]. Moreover, short-segment RNA tran-siently transcribed from the promoter upstream is also involved in the targeting of DNMTs to unique

loci, mediating the establishment of DNA methyla-tion [45]. In addition, DNA methylation localized in special ncRNAs sequences often participates in the regulation of these ncRNAs’ expression; for example, the methylation state of a downstream CpG island in the imprinted Airn regulates the transcription initia-tion, elongation and methylation status of its promoter region [46]. In addition, there is also an interdependent relationship between ncRNAs and histone modifi-cations. Similarly to DNA methylation, the histone modification status of ncRNA gene determines its expression [47]. Based on the observation that genes actively transcribed by RNA polymerase II are sym-bolized by trimethylation of lysine 4 of histone H3 (H3K4me3) at their promoter and trimethylation of lysine 36 of histone H3 (H3K36me3) along the length of the transcribed region, this distinctive structure referred to as a ‘K4–K36 domain’ is utilized to identify unknown functional noncoding RNAs, which leads to the discovery of lincRNA-COX2. what’s more, the genome-wide recognition of many ncRNAs, particu-larly lncRNAs, is based on the existence of specific his-tone modifications (e.g., H3K4me3 and H3K36me3) [47]. On the other hand, cis- or trans- actions of various ncRNAs is achieved by the interaction with specific histone modification enzymes that are targeted to spe-cific sites, resulting in the establishment of a specific histone modification state. A classic example is Hox antisense intergenic RNA (HOTAIR), which targets two specific histone modification enzymes (PRC2 [48] and LSD1 [49]) to the HOXD locus, to mediate silencing [50]. The positioning of the nucleosomes containing the key DNA regulatory element within the ncRNA gene promoter affects its transcription initiation, meantime, nucleosome distribution within the ncRNA gene body also regulates transcription elongation and splicing of the precursor. In addition, ncRNAs can also influence nucleosome position-ing through different mechanisms [51]. For example, PHO5 antisense ncRNA transcription itself can affect the stability and turnover of the nucleosome within the adjacent protein-coding gene PHO5, which pro-motes transcription [52]. Perhaps ncRNAs may also potentially recruit directly or indirectly various chro-matin remodeling complexes to specific target sites [53], similar to the mode of action of HOTAIR and HOXA transcript at the distal tip (HOTTIP) [54], which mediate changes in nucleosome positioning.

Crosstalk among various ncRNAsVarious intracellular ncRNAs form a mutual-com-munication ncRNA network by establishing exten-sive crosstalk that is achieved by the following several mechanisms:




• Regulatory lncRNA affects the sncRNAs pre-cursor’s mature processing. For example, the 800-nucleotide C. elegans lncRNA rncs-1 [55] that undergoes splicing and polyadenylation can trans-inhibit the generation of mature sncRNAs from other transcripts by competitively combining with Dicer or accessory double-stranded RNA-binding proteins;

• Regulatory lncRNA is the precursor of several functional ncRNAs; for example, many regulatory sncRNA-like sRNAs are produced from lncRNAs with an extended internal hairpin or two possess-ing complementary sequence lncRNAs. These sRNAs are then loaded into the RISC/RITS-like complexes to mediate silencing [56,57];

• Regulatory lncRNA can regulate the activity of sncRNAs by base pair formation between the lncRNAs and target sncRNAs, namely, the sponge effect. For example, during muscle differentiation, a specific lncRNA (linc-MD1) can ‘sponge’ miR-133 and miR-135, which regulate the expression of two key transcription factors – MAML1 and MEF2C [58];

• In terms of generation of sncRNA, there is a rela-tionship of cooperation or antagonism among various regulatory ncRNAs; for example, Arabi-dopsis miRNA mediates the mature processing of tasiRNA precursor transcript [59].

However, in Drosophila, the endo-siRNAs path-way may restrain the piRNA expression in the soma. In addition, these sncRNA pathways cooperate or antagonize each other at some level; for example, they may compete or share common substrates and effec-tor proteins, thus enabling these pathways to achieve crossregulation [60].

Epigenetics: the language of the cellOver the last 100 years, various modes of communi-cation have undergone extensive development, which has promoted the exchange of information around the world, finally resulting in the formation of a real plu-ralistic society. An obvious effect of extensive exchange is that the various resources on earth are rationally utilized on the basis of best interest, which greatly improves productivity and creates abundant mate-rial wealth. Accordingly, human society realizes this unprecedented achievement, which is largely attributed to the extensive exchange of information. In compari-son, since the first true cells appeared, various intracel-lular substructures, such as the nucleus and organelles, have not undergone significant changes. However, the

operating efficiency of the cell has undergone quali-tative changes (e.g., human cells vs Paramecium) so that the cell gradually adapts to the requirement of a complex environment and multicellular organismal development. It is still unclear whether this type of improvement in operating efficiency is also attributed to extensive intracellular communication. If so, what mediates this type of whole cell-wide information exchange? Moreover, in what ways?

The answer may be the real purpose for which an organism evolves with epigenetics, which enables the extensive exchange of information. In addition to the increasingly complex extracellular environment and formation of multicellular organisms, the highly effi-cient and synergistic cellular response among various cellular processes is the result of evolutionary pressure selection. On the basis of a limited genome, epigenetics organizes numerous activities in the cytoplasm and nucleus by taking various epigenetic markers as lan-guage to mediate extensive communication among different cellular processes, which finally resulted in genome-wide or whole cell-wide integral regulation. Thus, the cell generates an optimal response to various intracellular and extracellular stimuli.

However, in what ways do various epigenetic mark-ers mediate this extensive communication at the scale of the whole cell?

ncRNAs act as the language of the cellRecent confirmation of the ceRNA hypothesis indi-cates that various miRNAs act as the language for communication among mRNAs, synergizing the expression of various related proteins that are involved in different cellular processes. Emerging evidence has shown that various mRNAs exhibit multiple miRNA-binding sites, namely miRNA response elements, at the 3´ end. More than 60% of human protein-coding genes maintain their base pairs with miRNAs under natural selection pressure [61]. A single miRNA can also match with various mRNAs at the 3´ end, and target recognition is achieved by matching mRNA’s ∼7 nt site with the miRNA seed region [62]. However, the site-binding efficiency is affected not only by the seed match types but also by the characteristics surround-ing the site, which results in a single miRNA pos-sessing a gradient affinity to various target mRNAs, thereby achieving regulation of different degrees [63]. Thus, there are many-to-many relationships between miRNAs and mRNAs, and this type of relationship often initiates two synergistic processes:

• Coordinated process: a single miRNA can simul-taneously downregulate various proteins that are involved in different cellular processes, to achieve




a simultaneous response of various related cel-lular pathways induced by a specific stimulus (Figure 2A) [64];

• Antagonistic process: the quantity of intracellular specific miRNA is relatively limited (namely a cell type-specific ‘ncRNAs library’, however, this type

of ncRNA library is in a dynamic equilibrium, where specific ncRNAs generally change in quan-tity and quality in response to various intracel-lular and extracellular signals), new miRNAs are not only produced by Dicer processing or RdRPs, but various other mechanisms are used to inhibit miRNA production, thereby regulating the quan-



Figure 2. The special ncRNA (miRNA, endo-siRNA and piRNA) mediates synergistic expression (coordination and antagonism) among related genes at the post-transcriptional level. (A) The different expression level of special ncRNA will simultaneously mediate mRNA degradation of related genes (Gene A, B and C) by different degrees at the post-transcription level. (B) Based on the limited special ncRNA, the different expression level of Gene B will affect the expression of two other related genes (Genes A and C) by different degrees at the post-transcription level.

m7GGene A

State A

State A

State B

State B

State C

State C

Gene B Gene C

Gene A Gene B Gene C


m7GAAAA AAAA m7G AAAA

m7G

m7G

AAAA

AAAAm7G AAAA

m7G AAAAm7G AAAA

m7G AAAAm7G AAAA

AAAAAAAA

m7Gm7G

ncRNAs:basal

ncRNAs:best

Gene A:basal

Gene B:medium

Gene C:basal

ncRNAs:medium

Gene A:Basal

Gene B:Basal

Gene C:Basal

m7G m7GAAAA AAAA m7G AAAAm7G AAAAm7G

m7G

AAAA

AAAA

m7G AAAAm7G

m7G

AAAA

AAAA

AAAAAAAA

AAAA

m7Gm7G

m7G

m7G AAAAAAAAAAAA

m7Gm7G

m7G AAAAm7G AAAAm7G AAAA



m7G AAAAm7G AAAA


m7G AAAA


AAAAm7GAAAAm7G

AAAAm7GAAAAm7GAAAAm7G

m7G AAAAAAAAAAAA

m7Gm7G


m7G AAAAm7G AAAAm7G AAAAm7G AAAA


m7G AAAAm7G AAAAm7G AAAAm7G AAAA

AAAAm7GAAAAm7G


m7G

AAAA

AAAA

m7G AAAAm7G

m7G

AAAA

AAAA

AAAAAAAA

AAAA

m7Gm7G

m7G


m7G AAAAm7G AAAA

m7G AAAAm7G AAAA

AAAAm7GAAAAm7G

Gene A:Basal

Gene B:Best

Gene C:Basal


m7G

AAAA

AAAA

m7G AAAAm7G

m7G

AAAA

AAAA

AAAAAAAA

m7Gm7G



m7G AAAAm7G AAAAm7G AAAAm7G AAAAm7G AAAAm7G AAAA

mRNA A

mRNA B

mRNA C

Protein ASpecial ncRNA

Exosome

mRNAs’ affinity for the ncRNA:Gene C > Gene B > Gene A

Protein B

Protein C

i

iiii

iii

ii

ii

80 Epigenomics (2014) 6(1) future science group


Gene A

ncRNAs:basal ncRNAs:

medium

State A

Gene B Gene C Gene A

State B

Gene B Gene C

Gene A

Gene A:basal

Gene B:basal

Gene C:basal

Gene A:basal

Gene A:basal

Gene B:medium

Gene C:basal

State A

Gene B Gene C

Gene A

State B State C

Gene B Gene C

Gene B:best

Gene C:basal


ncRNAs:best

Gene A

State C

Gene B Gene C

m7G AAAAm7G AAAAAAAAm7Gm7G AAAAm7G AAAAAAAAm7Gm7G AAAAm7G AAAA

m7G AAAAm7G AAAAAAAAm7G

m7G AAAAm7G AAAAAAAAm7Gm7G AAAAm7G AAAAAAAAm7G

m7G AAAAAAAAm7Gm7G AAAAm7G AAAA

AAAAm7G m7G AAAAm7G AAAAm7G AAAA

Chromatin

Special-cellnascent RNA

Silencing machine

Special ncRNA

Transcriptional active marker

Transcriptional negative marker

Gene-special nascent RNA’s affinity for the ncRNA: Gene C > Gene B > Gene A

i

iiii

iii

ii

ii


tity of miRNAs. For example, the MCPIP1 nucle-ase inhibits the Dicer effect via cleavage of the terminal loops of pre-miRNAs, which terminates miRNA production [65].

Perhaps miRNA quantity is also potentially moni-tored at the transcriptional level. When various mRNAs compete with the same miRNAs, an antago-nistic relationship is formed among these mRNAs, namely ceRNAs (Figure 2B). A typical example is the PTEN-a key tumor suppressor, whose unbal-anced expression in tissue often results in tumorigen-esis. PTEN encodes a phosphatase, which catalyzes 3,4,5-triphosphoinositide into 4,5-diphosphoinositide, antagonizing the PI3K/Akt signal pathway and func-tions as a tumor suppressor. PTEN expression is sub-ject to miRNA regulation, such as miR-181, -200b, -25 and -92a. These limited miRNAs may also be com-bined with other PTEN ‘ceRNAs’ (i.e., the pseudogene lncRNA-PTENP1 and ZEB2). When this miRNA-associated equilibrium of mRNAs is broken (which is established under specific physical conditions), dis-ease occurs. For example, at the transcriptional level, when the expression of some PTEN ceRNAs is down-regulated, the quantity of miRNA-combined PTEN mRNAs increase, which downregulates the expression of PTEN, resulting removal of the inhibitory effect of the PI3K/Akt signaling pathways and, ultimately, tumorigenesis [66–68]. However, this type of antago-nistic effect does not always occur in pathological processes. We propose that the ‘ceRNA strategy’ may also be applied to specific developmental processes. For example, during immune system development, undifferentiated CD4+ T cells (Th0) develop into Th1, Th2, Th17 and Treg cells. When CD4+ T cells are in an undifferentiated state, specific miRNAs may simultaneously downregulate the mRNA expression of various developmental key genes that directional differentiations of various helper T cells require (e.g., Stat4, Gata3, Il17a-Il17f and Foxp3), thereby inhibit-ing differentiation (coordination). When the undif-ferentiated cells are subjected to specific stimuli under polarization conditions, this induces the upregulation of the expression of development-related key genes at the transcriptional level, as a result, mRNAs levels of special developmental key genes are greatly improved. Subsequently, the common inhibitory balance state is thus broken, which eventually results in specific direc-tional differentiation and simultaneously suppresses

other directional differentiation (antagonism). How-ever, we can often observe cell transdifferentiation between two types of differentiated cells, this reason may be that the equilibrium under some differenti-ated state is again broken. After it passes through the intermediate state, a new equilibrium under another differentiated state is established. Recent studies have revealed various immune system development-related miRNAs, such as miR-155, miR-21, miR-103 and miR-29 [69,70], especially miR-21, which has a cen-tral role in setting a balance between Th1 and Th2 responses to allergens [71].

Similarly to miRNA-mediated post-transcriptional regulation, this synergistic strategy may also be applied to miRNA-mediated transcriptional silencing regu-lation. Counterparts of genes, which are subjected to regulation by the same miRNAs, are cytoplasmic mRNAs. A ‘language’ consisting of special miRNAs mediates the synergistic expression of genome-wide related genes. It should be emphasized that protein-coding genes exhibit multiple copies in eukaryotic genomes. For example, humans have four HOX loci [72]. Similarly to cytoplasmic miRNAs, specific miRNAs can also simultaneously downregulate multiple target genes in the genome (coordination) (Figure 3A). More-over, the ‘ceRNA’ still exists (antagonism) and this type of ‘ceRNA’ relationship is formed among genes that are subjected to the same miRNA regulation and ‘sponge function’ lncRNAs, which specifically match with the miRNA (e.g., lnc-MD1 vs miR-133 and miR-135). When special gene expression level changes by different degrees, the ‘ceRNA’ phenomenon occurs (Figure 3B). This strategy of miRNA-mediated syner-gistic transcription expression among related genes is also shared with trans-acting lncRNAs. Importantly, regulation of miRNA-mediated silencing at the tran-scriptional and post-transcription levels is not indepen-dent of each other and may cooperate with each other to achieve the most optimal match (Figure 3).

Nearly all of the regulatory ncRNAs (more than the miRNAs) may adopt a similar mode of action, by which ncRNAs mediate the extensive exchange of information that acts as the languagee of communi-cation. For example, piRNAs can act as the language to synergize cellular processes in the cytoplasm and nucleus, and to maintain germline genome stabil-ity and its germline property [10]. Endo-siRNAs also utilize this language ability to completely resist patho-genic infections or endogenous parasitic sequence reac-



Figure 3. The special ncRNA (miRNA, endo-siRNA, piRNA) mediates synergistic expression (coordination and antagonism) among related genes at the transcriptional level (see facing page). (A) The different expression level of special ncRNA will simultaneously mediate transcriptional silencing of related genes (Genes A, B and C) by different degrees at the transcription level. (B) Based on the limited special ncRNA, the different expression level of Gene B will affect the transcription of two other related genes (Genes A and C) by different degrees at the transcription level.


tivation at the transcriptional and post-transcriptional levels [60]. In addition, trans-acting lncRNAs not only exert their function by acting as ceRNAs, but they also act via the communication language to mediate synergistic activity through a miRNA-like mode of action at the transcriptional and post-transcriptional level. For example, lncRNA 1/2-sbsRNA is involved in cytoplasm-specific mRNA degradation, while trans-acting lncRNAs mediate related gene transcriptional silencing (e.g., lncRNA-P21 induced by P53 medi-ates various gene trans-silencing during p53-mediated responses [73]). Furthermore, lncRNAs also achieve the extensive exchange of information by establish-ing specific intracellular substructures (e.g., MEN ε/β and Xlsirts are involved in the establishment of nuclear paraspeckles and organization of the cytoskeleton, respectively) [74,75] and by processing other RNAs.

Covalent modifying enzymes & remodeling complexes act as the language of the cellIn what ways do other epigenetic markers mediate the extensive exchange of information? What is the language for this communication? In terms of DNA methylation, histone modification and nucleosome positioning, their genome-wide epigenetic patterns are mainly established through the activity of various spe-cific enzymes (nucleosome positioning is partly attrib-uted to the DNA sequence itself [76]). Thus, for this type of extensive communication mediated by these epigenetic markers, various specific enzymes comprise the language for communication (e.g., WDR5-MLL [77,78]), and DNA methylation, histone modification or nucleosome positioning at special loci are the even-tual results of interaction among mobile modification enzymes or remodeling complexes, which is similar to mobile ncRNAs. Usually, these covalent modify-ing enzymes or remodeling complexes are the com-ponents of larger protein complexes or interact with other chromatin-binding factors (e.g., transcription factors or ncRNAs), other members of these protein complexes or interacted-chromatin binding factors can determine their target specificity, enabling the target-ing of special enzymes to various protein-coding gene loci or control enzyme activities. Finally, these modi-fying enzymes or remodeling complexes are targeted to various related protein-coding genes loci, which simultaneously establish the specific epigenetic envi-ronment, known as ‘epigenetic codes’, of each gene in the related genetic group, thereby achieving the coor-dinated expression of the entire genetic group (coor-dination). For example, when undifferentiated CD4+ T cells perform directional differentiation under Th17 cell polarization conditions, at certain Th17-specific genes promoter loci (Il17a-Il17f and RORγt), the active

marker H3K4me3 significantly increases. In addition, the inhibitory marker H3K27me3 decreases. Compar-atively, the Th1- and Th2-specific gene promoter loci (IFN-γ, IL-4, IL-13 and T-box21, Gata3) the opposite situation occurs, where H3K27me3 levels are upregu-lated, while H3K4me3 levels are downregulated. The final result is that under Th17 polarization conditions, the Th17-specific developmental genes expression is enhanced, thereby promoting the establishment of Th17-specific biological characteristics. Moreover, the Th1- and Th2-developmental gene expression is further downregulated, which inhibits Th17-opposite Th1 and Th2 directional differentiation [79]. During directional differentiation, distinct expression pat-terns of the different genetic groups may be established through the following mechanism: under Th17 cell polarization conditions, the cell first upregulates the expression of Th17-specific transcription factors or ncRNAs, which mediate the targeting of mixed lineage leukemia (MLL) protein family members to develop-mental key genes whose expression is required by Th17 differentiation (it should be emphasized that the quan-tity of the intracellular active or inhibitory machinery may be relatively limited). The large protein complex consisted of MLL members often contains another enzyme that can catalyze the removal of inhibitory marker H3K27me3 (e.g., UTX [80]), which can simul-taneously achieve the upregulation of H3K4me3 and the downregulation of H3K27me3, and rapidly switch a specific gene’s expression state.

However, in an undifferentiated state, various devel-opmental key genes promoter loci often form bivalent chromatin structures, which consists of an active and inhibitory marker, H3K4me3 and H3K27me3, respec-tively, in a type of transcription-poised state. This may be the concurrent joint effect of two relatively bal-anced enzymes, MLL and EZH2 (EZH2 catalyzes the formation of H3K27me3) (Figure 4A). During Th17 directional differentiation, this balance is broken, in which the majority of the MLL is targeted to vari-ous Th17-specific developmental gene loci. However, the remaining EZH2 is mainly localized to other key developmental genes, whose expression is required by various Th17-opposite directional differentiations; finally, a new balance is established. Interestingly, this ‘ceRNA-like’ strategy is also applied in this context. At this point, the Th1- and Th2-specific developmental genes are the ‘ceRNAs’ of the Th17-specific develop-mental genes (antagonism) (Figure 4B). However, this newly built balance is relative; when the differentiated cells are subjected to certain stimulus, the established balance is again broken (Figure 4C). For example, when the Th17 cell is stimulated by IL-12, the balance tend-ing to induce Th17 directional differentiation is par-




tially broken. At the IFN-γ gene locus, the H3K4 me3 level is partly upregulated and, simultaneously, the H3K27 me3 levels are also partially downregulated, while the IL-17A gene locus occurs the opposite situa-tion. The total result is upregulation of IFN-γ expres-sion and downregulation of IL-17A expression. In this case, IL-12 stimulation induces a slight upregulation of Th1-specific transcription factors or ncRNAs expres-sion level, which competitively bind MLL to Th1-spe-cific genes (e.g., IFN-γ). Accordingly, because of the loss of MLL, the partial Th17-specific genes expres-sion is downregulated (e.g., IL-17A) [81]. Perhaps it is the molecular mechanism that underlies the founda-tion of plasticity in CD4+ T-cell lineage differentiation [82,83]. It is that epigenetic markers, H3K27me3 and H3K4me3, perform an important role during inducing the plasticity of CD4+ T cell lineage. However, the other extreme lies in the transdifferentiation process [84]. In this case, one balance established under a differenti-ated state is thoroughly broken, while another balance tending to induce another type of directional differen-tiation is completely established (Figure 4B, C & D) [85]. The same strategy may also play an important role in early embryonic development [86,87] and multipotent stem cell directional differentiation [88–90] because a large number of key developmental genes also possess a bivalent chromatin structure.

The strategy that histone modification enzymes func-tioning as a language of communication to mediate the formation of a synergistic relationship of expression (coordination and antagonism) among related genetic groups may also be applied to nucleosome positioning and DNA methylation. For nucleosome positioning, chromatin remodeling enzymes catalyzing the exposure or coverage of the key DNA-binding sites are the ‘lan-guage’ for communication among the related genes [91]. The relationship of ‘ceRNAs’ or coordination is formed among the key position nucleosomes that contain impor-tant regulatory elements, thus mediating extensive com-munication among the genome-wide protein-coding genes or various activities that occur in chromatin (e.g., between nucleosome positioning involved in DNA dam-age repair and nucleosome positioning involved in DNA replication). However, in the context of DNA methyla-tion, it is similar to histone modifications.

Interestingly, these modifying enzymes catalyze not only histones but also a large number of nonhistone proteins. Various nonhistone proteins can undergo dif-ferent types of post-translational modifications, such as phosphorylation and dephosphorylation, acetyla-tion and deacetylation, methylation and demethyl-ation, ubiquitination, and sumoylation and its removal, among others. Importantly, these nonhistone proteins are subjected to the same set of enzymes that act on the

histone. A typical example is lysine acetylation, which occurs in more than 80 transcription factors, many nuclear regulators and various cytoplasmic proteins. Acetylation not only regulates different activities within the nucleus, but it is also involved in the regulation of various cytoplasmic processes, including the regulation of α-tubulin acetylation, which is involved in cytoplas-mic skeletal dynamics and assembly, the regulation of P53 acetylation is involved in the DNA damage repair response pathway, and the regulation of Foxo acetyla-tion in the IGF signal pathway [92]. In addition, the AMPK catalytic subunit is subjected to the dynamic regulation of P300 (acetylation) and HDAC1 (deacet-ylation), which are involved in the AMPK signaling pathway [93]. A variety of acetylation activities in the cytoplasm and nucleus are all subject to the same set of acetyltransferase and deacetylase activities. Similar activities also occur with other modifications. In these cases, functioning as a language for extensive commu-nication, modifying enzymes not only participate in the regulation of various cytoplasmic and nuclear activities but also mediate extensive communication between cytoplasmic and nuclear activities, which mainly occur at the chromatin level and results in the whole-cell-wide integral regulation of various cellular processes, which greatly improves the cell operating efficiency.

Taken together, various epigenetic markers not only participate in the regulation of different cellular pro-cesses by themselves, but organisms have also evolved with a unique function of epigenetics, namely that epi-genetics is involved in mediating the extensive infor-mation exchange within the entire cell by acting as a language of communication, which mediates integral regulation among various cellular processes. This ulti-mately results in the cell generating an optimal response to different intracellular and extracellular stimulations.

Importantly, no matter whether the communica-tional ‘language’ is consisted of various regulatory ncRNAs or various modifying enzymes or remodeling enzymes, they all perform extensive crosstalk longitu-dinally and laterally. Various ncRNAs, modification enzymes and remodeling enzymes often collaborate with each other to fine-tune various biological events in a cell’s response to stimuli, finally the cell makes the optimal reply to the variation of extracellular or intra-cellular environment. For example, a mRNA 3´ end has multiple MERs; various modifying enzymes can simultaneously act on this specific promoter locus [94] and trans-acting ncRNA may simultaneously mediate the establishment of histone modifications and DNA methylation patterns at this particular promoter locus. The cell utilizes various epigenetic markers to weave into this complex language system to mediate extensive intracellular communication.




Conclusion & future perspectiveEpigenetic research has become one of the largest forefront research hotspots in biological research. In addition to abundant high-technological development, such as the genome-wide chromatin immunoprecipi-tation-deep sequencing (ChIP-Seq) technology and high-resolution genome-wide mapping technologies,

researchers may conduct epigenetics research of not only a single gene or specific chromosome region, but rather, the whole genome, to understand their general distributional law (particularly for DNA methylation [95], histone modifications [96] and nucleosome posi-tioning [97]). In addition, the discovery of different epigenetic machinery also greatly promotes epigenetic



Figure 4. DNA methylation, histone modification and nuclesome position mediate the extensive communication among related genes groups, achieving coordinated or antagonistic expression. (A) Under the undifferentiated state, the relative balance between activating machine and silencing machine promotes the formation of a poised state (simultaneously possess active and inhibitory epigenetic markers) in each gene of various related genes groups. (B) Under the differentiated State A (e.g., the Th17 polarization condition), the relative balance is broken, in which Gene group 1 special TFs target active machinery to each gene of Gene group 1, promoting transcription, however, remaining silencing machine is stably bound to Gene group 2, mediating silencing. (C) When the differentiated State A cell is slightly stimulated by the differentiated State B polarization condition (e.g., IL-12), the established balance is again broken, a little of Gene group 2 special TFs target active machine to some genes of Gene group 2. However, because of active machinery’s removal, the partial genes of Gene group 1 are transcriptional silencing. (D) When the differentiated State A cell is completely stimulated by the differentiated State B polarization condition, another balance state is established and the transdifferentiation occurs; finally the differentiated State B cell is formed. TF: Transcription factor.

Gene A

Gene E Gene F Gene G Gene H Gene E Gene F Gene G Gene H

Gene group 1: Gene A, B, C, D

Gene group 2: Gene E, F, G, H

State A: the undifferentiated



State B: the differentiated State A

Gene B Gene C Gene D

Gene A

Gene E Gene F Gene G Gene H



Activating machine (promote the removal of DNA methylation or the establishmentof active histone modification or nucleosome position activating transcription)

Sillencing machine (promote the establishment of DNA methylation or the inhibitoryof active histone modification or nucleosome position inhibiting transcription)

State C: the differentiated State A

Gene E Gene F Gene G Gene H



Transcriptional active markers

Transcriptional negative markers

Gene group 1 special TF

Gene group 2 special TF

State D: the differentiated State B

Gene B

B

Gene C Gene D Gene A Gene B Gene C Gene D

Gene A Gene B Gene C Gene D


research. Thus far, researchers have gradually revealed novel modes of action of different epigenetic markers and demonstrated the presence of extensive cross-talk among various epigenetic markers, which are involved in weaving into so-called ‘epigenetic codes’ that fine-tunes different cellular processes. Epigenetic markers, particularly epigenetic markers based on chromatin, not only enrich biological genetic information, but fine-tune various activities, generating an optimal cel-lular response to various extracellular and intracellu-lar stimuli. However, there are still many challenges, which require further research: how do these individ-ual epigenetic marker ‘alphabet’ combine to spell func-tional ‘words’? By which molecular mechanism does these functional ‘words’ instruct the exact expression of different protein-encoding and ncRNAs genes? By which mechanism do various epigenetic marker pat-terns continue to maintain after cell division? How does the site-specific epigenetic state reverse into dif-ferent epigenetic states during cell differentiation and pathogenesis?

With the development of various sequencing technol-ogies, a large number of functional ncRNAs have been uncovered and have been shown to participate in various cellular processes. Researchers are gradually becoming aware of the striking effects of ncRNAs in the cell, par-ticularly various functionally diverse lncRNAs. These regulative ncRNAs are also passed into daughter cells after cell division to continue exerting their function. They may be indirectly inherited through the establish-ment of specific chromatin epigenetic environments on the ncRNA loci, forming a stable transcriptional state. Thus, many ncRNAs, specifically regulatory ncRNAs, are also epigenetic markers. Undoubtedly, because of their diversity and abundance, various ncRNAs will grow in the epigenetic research field.

Although there are a large number of epigenetic markers in the cytoplasm and in chromatin, these markers tend to regulate the occurrence of an indi-vidual event, such as the specific miRNA-mediated mRNA degradation. However, it remains unclear whether these abundant epigenetic markers can also mediate the integral regulation among various cellu-lar processes at the scope of the whole cell, in respond

to various intracellular and extracellular stimuli and organism developmental needs. For example, the expression of related-protein mRNAs or genes is con-certedly controlled during cell differentiation (e.g., some related genes’ expression are simultaneously upregulated and downregulated, or some are upregu-lated while another is downregulated). Recently the confirmation of the ceRNA hypothesis has provided a clue: perhaps all epigenetic markers, more than the miRNAs, may potentially possess a similar ‘language’ function to mediate extensive communication among various related protein-coding genes or in various cel-lular processes, to achieve integral regulation. The ‘epigenetic language’ may mediate extensive commu-nication among various metabolic pathways or signal-ing pathways to achieve overall regulation. In addition, there is extensive crosstalk between ncRNAs and chro-matin covalent modifications and, thus, different types of epigenetic languages may perform lateral commu-nication, which eventually forms a complex informa-tion-exchange network. Undoubtedly, this will greatly improve the cell’s operating efficiency. Language ability may be an important windfall of epigenetics, which is the flower of life. However, numerous studies are required to further elucidate ‘epigenetic language’-mediated synergistic regulation, particularly, that related to the ‘ceRNA’ phenomenon during normal organismal development or pathogenesis, and deeply define molecular mechanisms of this type of epigenetic language-mediated communication. Epigenetics will provide a new perspective for the mystery of life.

Financial & competing interests disclosureThis work is supported by the Ministry of Science and

Technology of China through Grants No 2012CB932503

and 2011CB933100; the National Natural Science Founda-

tion of China through Grants No 91029705, 81172864 and

81272317. The authors have no other relevant affiliations or

financial involvement with any organization or entity with a

financial interest in or financial conflict with the subject mat-

ter or materials discussed in the manuscript apart from those

disclosed.

No writing assistance was utilized in the production of this

manuscript.



Executive summary

ncRNAs: an important part of the epigenetics• ncRNA is not only involved in the establishment of the epigenetic patterns of various epigenetic markers, but ncRNAs are also an

important type of epigenetic marker.Extensive crosstalk among various epigenetic markers• There is extensive crosstalk among various epigenetic markers, including noncoding RNAs and other epigenetic markers.Epigenetics: the language of the cell• The language property of epigenetics is shared with ncRNAs and other epigenetic markers, in which ncRNAs themselves, covalent

modifying enzymes and remodeling complexes act as the language.


ReferencesPapers of special note have been highlighted as: l of interest l l of considerable interest

1 Gartler SM, Riggs AD. Mammalian X-chromosome inactivation. Ann. Rev. Genet. 17(1), 155–190 (1983).

2 Brink RA. Paramutation. Ann. Rev. Genet. 7(1), 129–152 (1973).

3 Wakimoto BT. Beyond the nucleosome: epigenetic minireview aspects of position–effect variegation in Drosophila. Cell 93, 321–324 (1998).

4 Baker BS, Gorman M, Marin I. Dosage compensation in Drosophila. Ann. Rev. Genet. 28(1), 491–521 (1994).

5 Bartolomei MS, Tilghman SM. Genomic imprinting in mammals. Ann. Rev. Genet. 31(1), 493–525 (1997).

6 Horwich AL, Weissman JS. Deadly conformations-protein misfolding in prion disease. Cell 89, 499–510 (1997).

7 Waddington C. The epigenotype. Endeavour 1, 18–20 (1942).

8 Goldberg AD, Allis CD, Bernstein E. Epigenetics: a landscape takes shape. Cell 128(4), 635 (2007).

9 Allis CD. Epigenetics. CSHL Press, NY, USA (2007).

10 Daxinger L, Whitelaw E. Understanding transgenerational epigenetic inheritance via the gametes in mammals. Nat. Rev. Genet. 13(3), 153–162 (2012).

l l Provides examples to demonstrate that certain development-related ncRNAs can be the transgenerational epigenetic markers via gametes.

11 Colby DW, Prusiner SB. Prions. Cold Spring Harb. Perspect. Biol. doi:10.1101/cshperspect.a006833 (2011).

12 Hannon GJ. RNA interference. Nature 418(6894), 244–251 (2002).

13 Bonasio R, Tu S, Reinberg D. Molecular signals of epigenetic states. Sci. Signal. 330(6004), 612 (2010).

14 Alabert C, Groth A. Chromatin replication and epigenome maintenance. Nat. Rev. Mol. Cell Biol. 13(3), 153–167 (2012).

15 Sarkies P, Sale JE. Cellular epigenetic stability and cancer. Trends Genet. 28(3), 118–127 (2012).

16 Margueron R, Reinberg D. Chromatin structure and the inheritance of epigenetic information. Nat. Rev. Genet. 11(4), 285–296 (2010).

17 Rajasethupathy P, Antonov I, Sheridan R et al. A role for neuronal piRNAs in the epigenetic control of memory-related synaptic plasticity. Cell 149(3), 693–707 (2012).

18 Hou F, Sun L, Zheng H, Skaug B, Jiang QX, Chen ZJ. MAVS forms functional prion-like aggregates to activate and propagate antiviral innate immune response. Cell 146(3), 448–461 (2011).

19 Soto C. Transmissible proteins: Expanding the prion heresy. Cell 149(5), 968–977 (2012).

20 Pastor WA, Pape UJ, Huang Y et al. Genome-wide mapping of 5-hydroxymethylcytosine in embryonic stem cells. Nature 473(7347), 394–397 (2011).

21 Ito S, D’alessio AC, Taranova OV, Hong K, Sowers LC, Zhang Y. Role of Tet proteins in 5mC to 5hmC conversion, ES-cell self-renewal and inner cell mass specification. Nature 466(7310), 1129–1133 (2010).

22 Talbert PB, Henikoff S. Histone variants – ancient wrap artists of the epigenome. Nat. Rev. Mol. Cell Biol. 11(4), 264–275 (2010).

23 Zhu J, Yamane H, Paul WE. Differentiation of effector CD4 T cell populations*. Ann. Rev. Immunol. 28, 445–489 (2009).

24 Salmena L, Poliseno L, Tay Y, Kats L, Pandolfi PP. A ceRNA hypothesis: the Rosetta stone of a hidden RNA language? Cell 146(3), 353–358 (2011).

l l First paper to propose the ceRNA hypothesis.

25 Rubio-Somoza I, Weigel D, Franco-Zorilla JM, García JA, Paz-Ares J. ceRNAs: miRNA target mimic mimics. Cell 147(7), 1431 (2011).

26 Sumazin P, Yang X, Chiu HS et al. An extensive microRNA-mediated network of RNA-RNA interactions regulates established oncogenic pathways in glioblastoma. Cell 147(2), 370–381 (2011).

27 Iyer VR. Nucleosome positioning: bringing order to the eukaryotic genome. Trends Cell Biol. 22(5), 250–256 (2012).

28 Jiang C, Pugh BF. Nucleosome positioning and gene regulation: advances through genomics. Nat. Rev. Genet. 10(3), 161–172 (2009).

29 Bannister AJ, Kouzarides T. Regulation of chromatin by histone modifications. Cell Res. 21(3), 381–395 (2011).

30 Portela A, Esteller M. Epigenetic modifications and human disease. Nat. Biotechnol. 28(10), 1057–1068 (2010).

31 Kouzarides T. Chromatin modifications and their function. Cell 128(4), 693 (2007).

32 Esteller M. Non-coding RNAs in human disease. Nat. Rev. Genet. 12(12), 861–874 (2011).

33 Mattick JS. The genetic signatures of noncoding RNAs. PLoS Genet. 5(4), e1000459 (2009).

34 Rinn JL, Kertesz M, Wang JK et al. Functional demarcation of active and silent chromatin domains in human HOX loci by non-coding RNAs. Cell 129(7), 1311–1323 (2007).

35 Aravin AA, Hannon GJ, Brennecke J. The Piwi-piRNA pathway provides an adaptive defense in the transposon arms race. Science 318(5851), 761–764 (2007).

36 Suter CM, Martin DIK. Paramutation: the tip of an epigenetic iceberg? Trends Genet. 26(1), 9–14 (2010).

37 Hollick JB. Paramutation and development. Ann. Rev. Cell Dev. Biol. 26, 557–579 (2010).

38 Chandler VL. Paramutation: from maize to mice. Cell 128(4), 641–645 (2007).

39 Erhard KF, Hollick JB. Paramutation: a process for acquiring trans-generational regulatory states. Curr. Opin Plant Biol. 14(2), 210–216 (2011).

40 Girard A, Sachidanandam R, Hannon GJ, Carmell MA. A germline-specific class of small RNAs binds mammalian Piwi proteins. Nature 442(7099), 199–202 (2006).

41 Wei W, Ba Z, Gao M et al. A role for small RNAs in DNA double-strand break repair. Cell 149(1), 101–112 (2012).




42 Kritikou E. Small RNAs: endo-siRNAs truly endogenous. Nat. Rev. Mol. Cell Biol. 9(6), 426–427 (2008).

43 Hung T, Wang Y, Lin MF et al. Extensive and coordinated transcription of noncoding RNAs within cell-cycle promoters. Nat. Genet. 43(7), 621–629 (2011).

44 Zhang H, Zhu JK. RNA-directed DNA methylation. Curr. Opin Plant Biol. 14(2), 142–147 (2011).

45 Han J, Kim D, Morris KV. Promoter-associated RNA is required for RNA-directed transcriptional gene silencing in human cells. Proc. Natl Acad. Sci. USA 104(30), 12422–12427 (2007).

46 Koerner MV, Pauler FM, Hudson QJ et al. A downstream CpG island controls transcript initiation and elongation and the methylation state of the imprinted Airn macro ncRNA promoter. PLoS Genet. 8(3), e1002540 (2012).

47 Mitchell Guttman IA, Garber M, French C et al. Chromatin signature reveals over a thousand highly conserved large non-coding RNAs in mammals. Nature 458(7235), 223–227 (2009).

48 Margueron R, Reinberg D. The Polycomb complex PRC2 and its mark in life. Nature 469(7330), 343–349 (2011).

49 Shi Y, Lan F, Matson C et al. Histone demethylation mediated by the nuclear amine oxidase homolog LSD1. Cell 119(7), 941–953 (2004).

50 Tsai MC, Manor O, Wan Y et al. Long noncoding RNA as modular scaffold of histone modification complexes. Science 329(5992), 689–693 (2010).

51 Hainer SJ, Pruneski JA, Mitchell RD, Monteverde RM, Martens JA. Intergenic transcription causes repression by directing nucleosome assembly. Genes Dev. 25(1), 29–40 (2011).

52 Wilusz JE, Sunwoo H, Spector DL. Long noncoding RNAs: functional surprises from the RNA world. Genes Dev. 23(13), 1494–1504 (2009).

l Systematically summarizes various functional long noncoding RNAs.

53 Zhou Y, Schmitz KM, Mayer C, Yuan X, Akhtar A, Grummt I. Reversible acetylation of the chromatin remodelling complex NoRC is required for non-coding RNA-dependent silencing. Nat. Cell Biol. 11(8), 1010–1016 (2009).

54 Wang KC, Yang YW, Liu B et al. A long noncoding RNA maintains active chromatin to coordinate homeotic gene expression. Nature 472(7341), 120–124 (2011).

55 Hellwig S, Bass BL. A starvation-induced noncoding RNA modulates expression of Dicer-regulated genes. Sci. Signal. 105(35), 12897 (2008).

56 Mercer TR, Dinger ME, Mattick JS. Long non-coding RNAs: insights into functions. Nat. Rev. Genet. 10(3), 155–159 (2009).

57 Ogawa Y, Sun BK, Lee JT. Intersection of the RNA interference and X-inactivation pathways. Science 320(5881), 1336–1341 (2008).

58 Cesana M, Cacchiarelli D, Legnini I et al. A long noncoding RNA controls muscle differentiation by functioning as a competing endogenous RNA. Cell 147(2), 358–369 (2011).

l First paper to discover a special lncRNA that can regulate gene expression by sponge function.

59 Allen E, Xie Z, Gustafson AM, Carrington JC. microRNA-directed phasing during trans-acting siRNA biogenesis in plants. Cell 121(2), 207 (2005).

60 Ghildiyal M, Zamore PD. Small silencing RNAs: an expanding universe. Nat. Rev. Genet. 10(2), 94–108 (2009).

61 Friedman RC, Farh KKH, Burge CB, Bartel DP. Most mammalian mRNAs are conserved targets of microRNAs. Genome Res. 19(1), 92–105 (2009).

62 Bartel DP. MicroRNAs: target recognition and regulatory functions. Cell 136(2), 215–233 (2009).

63 Grimson A, Farh KKH, Johnston WK, Garrett-Engele P, Lim LP, Bartel DP. MicroRNA targeting specificity in mammals: determinants beyond seed pairing. Mol. Cell 27(1), 91–105 (2007).

64 Baek D, Villén J, Shin C, Camargo FD, Gygi SP, Bartel DP. The impact of microRNAs on protein output. Nature 455(7209), 64–71 (2008).

65 Suzuki HI, Arase M, Matsuyama H et al. MCPIP1 ribonuclease antagonizes dicer and terminates microRNA biogenesis through precursor microRNA degradation. Mol. Cell 44(3), 424–436 (2011).

66 Poliseno L, Salmena L, Zhang J, Carver B, Haveman WJ, Pandolfi PP. A coding-independent function of gene and pseudogene mRNAs regulates tumour biology. Nature 465(7301), 1033–1038 (2010).

67 Tay Y, Kats L, Salmena L et al. Coding-independent regulation of the tumor suppressor PTEN by competing endogenous mRNAs. Cell 147(2), 344–357 (2011).

l Display the importance of PTEN-related ceRNA during tumorigenesis.

68 Karreth FA, Tay Y, Perna D et al. In vivo identification of tumor-suppressive PTEN ceRNAs in an oncogenic BRAF-induced mouse model of melanoma. Cell 147(2), 382–395 (2011).

l l Provides a representative example to demonstrate the ceRNA hypothesis.

69 Xiao C, Rajewsky K. MicroRNA control in the immune system: basic principles. Cell 136(1), 26–36 (2009).

70 Muljo SA, Ansel KM, Kanellopoulou C, Livingston DM, Rao A, Rajewsky K. Aberrant T cell differentiation in the absence of Dicer. J. Exp. Med. 202(2), 261–269 (2005).

71 Lu TX, Hartner J, Lim E-J et al. MicroRNA-21 limits in vivo immune response-mediated activation of the IL-12/IFN-γ pathway, Th1 polarization, and the severity of delayed-type hypersensitivity. J. Immunol. 187(6), 3362–3373 (2011).

72 Shah N, Sukumar S. The Hox genes and their roles in oncogenesis. Nat. Rev. Cancer 10(5), 361–371 (2010).

73 Huarte M, Guttman M, Feldser D et al. A large intergenic noncoding RNA induced by p53 mediates global gene repression in the p53 response. Cell 142(3), 409–419 (2010).

74 Sunwoo H, Dinger ME, Wilusz JE, Amaral PP, Mattick JS, Spector DL. MEN ε/β nuclear-retained non-coding RNAs are up-regulated upon muscle differentiation and are




essential components of paraspeckles. Genome Res. 19(3), 347–359 (2009).

75 Shahbabian K, Chartrand P. Control of cytoplasmic mRNA localization. Cell. Mol. Life Sci. 1–18 (2012).

76 Segal E, Fondufe-Mittendorf Y, Chen L et al. A genomic code for nucleosome positioning. Nature 442(7104), 772–778 (2006).

77 Trievel RC, Shilatifard A. WDR5, a complexed protein. Nat. Struct. Mol. Biol. 16(7), 678–680 (2009).

78 Hess JL. MLL: a histone methyltransferase disrupted in leukemia. Trends Mol. Med. 10(10), 500–507 (2004).

l l Identifies the epigenetic instability of CD4+ T cells, which can be well explained with the language property of the epigenetics.

79 Wei G, Wei L, Zhu J et al. Global mapping of H3K4me3 and H3K27me3 reveals specificity and plasticity in lineage fate determination of differentiating CD4+ T cells. Immunity 30(1), 155–167 (2009).

80 Lee MG, Villa R, Trojer P et al. Demethylation of H3K27 regulates polycomb recruitment and H2A ubiquitination. Sci. Signal. 318(5849), 447 (2007).

81 Mukasa R, Balasubramani A, Lee YK et al. Epigenetic instability of cytokine and transcription factor gene loci underlies plasticity of the T helper 17 cell lineage. Immunity 32(5), 616–627 (2010).

82 Zhou L, Chong M, Littman DR. Plasticity of CD4+ T cell lineage differentiation. Immunity 30(5), 646–655 (2009).

83 Nakayamada S, Takahashi H, Kanno Y, O’Shea JJ. Helper T cell diversity and plasticity. Curr. Opin. Immunol. 24(3) 297–302 (2012).

84 Barrero MJ, Belmonte JCI. Regenerating the epigenome. EMBO Reports 12(3), 208–215 (2011).

85 Kim J, Efe JA, Zhu S et al. Direct reprogramming of mouse fibroblasts to neural progenitors. Proc. Natl Acad. Sci. USA 108(19), 7838–7843 (2011).

86 Vastenhouw NL, Schier AF. Bivalent histone modifications in early embryogenesis. Curr. Opin. Cell Biol. 24(3), 374–386 (2012).

87 Bernstein BE, Mikkelsen TS, Xie X et al. A bivalent chromatin structure marks key developmental genes in embryonic stem cells. Cell 125(2), 315–326 (2006).

88 Juliandi B, Abematsu M, Nakashima K. Chromatin remodeling in neural stem cell differentiation. Curr. Opin Neurobiol. 20(4), 408–415 (2010).

89 Cui K, Zang C, Roh TY et al. Chromatin signatures in multipotent human hematopoietic stem cells indicate the fate of bivalent genes during differentiation. Cell Stem Cell 4(1), 80–93 (2009).

90 Weng N, Araki Y, Subedi K. The molecular basis of the memory T cell response: differential gene expression and its epigenetic regulation. Nat. Rev. Immunol. 12(4), 306–315 (2012).

l l Shows that HDAC1 and p300 mediate the extensive crosstalk between cytoplasm and nucleus.

91 Xie W, Ling T, Zhou Y et al. The chromatin remodeling complex NuRD establishes the poised state of rRNA genes characterized by bivalent histone modifications and altered nucleosome positions. Proc. Natl Acad. Sci. USA 109(21), 8161–8166 (2012).

92 Yang XJ, Seto E. Lysine acetylation: codified crosstalk with other posttranslational modifications. Mol. Cell 31(4), 449–461 (2008).

93 Lin Y, Kiihl S, Suhail Y et al. Functional dissection of lysine deacetylases reveals that HDAC1 and p300 regulate AMPK. Nature 482(7384), 251–255 (2012).

94 Lee JS, Smith E, Shilatifard A. The language of histone crosstalk. Cell 142(5), 682–685 (2010).

95 Meissner A, Mikkelsen TS, Gu H et al. Genome-scale DNA methylation maps of pluripotent and differentiated cells. Nature 454(7205), 766–770 (2008).

96 Schones DE, Zhao K. Genome-wide approaches to studying chromatin modifications. Nat. Rev. Genet. 9(3), 179–191 (2008).

97 Brogaard K, Xi L, Wang JP, Widom J. A map of nucleosome positions in yeast at base-pair resolution. Nature 486(7404), 496–501 (2012).



Documents

Perspective Epigenetics: the language of the cell? · 2014-10-03 · may functionally acquire such a unique language characteristic. In summary, ... processes within the cell and