Epigenomic Explanations for the Uncertainty of Cancer Biomarkers

C l i n M e d International Library Citation: Shou-Tung C, Chia-Chen H, Yu-Wei L, Shu-Huei H (2016) Epigenomic Explanations for the Uncertainty of Cancer Biomarkers. Int J Pathol Clin Res 2:032 Received: February 10, 2016: Accepted: March 28, 2016: Published: April 01, 2016 Copyright: © 2016 Shou-Tung C, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Shou-Tung et al. Int J Pathol Clin Res 2016, 2:032


ISSN: 2469-5807
Volume 2 | Issue 2 lineage or tissue contain identical genomic material has faded away [2].Tumorigenesis occurs as a series of minor mutations rather than resulting from a single, high frequency mutation like P53 and APC [3].Moreover, in some cases mere "misfortune" can be credited as one of the reasons that a patient develops cancer [4,5].In efforts to explain at least some of the apparent stochasticity in tumorigenesis, there are recent attempts to identify cancer targets or markers associated with cancer.An epigenomic explanation for some of the unpredictability in cancer biomarkers might assist researchers and clinicians in the predictive use of these markers.

Clonal Evolution of Somatic Stem Cells and Cancers
Somatic cells were once believed to possess identical genomic contents since they are all derived from a single cell [6,7].During development, growth factors and other environmental signals reshape gene expression to generate different lineages of cells that later become specific tissues [8][9][10].During this process, external signals are relayed into the cells and are memorized as an inheritable code in the form of specific epigenetic modifications that restrain cell fate [11][12][13] (Figure 1).The internalization and translation of external signals into epigenetic codes occurs during the early stages of development and germ cell formation but rarely are identified or discussed during somatic stem cell differentiation or tumorigenesis [13][14][15].
In addition to determination of gene expression patterns, epigenetic modifications are also responsible for genome stability [32,33].Forty-five to fifty percent of the human genome consists

Introduction
With the accumulation of "big data" on genetics in human diseases [1], the likelihood of onset of various cancers is better understood.Identification of causal or associated changes in biomarkers would be impossible if the onset of cancer were entirely stochastic.On the other hand, big data has revealed a degree of uncertainty in the human genome.Even the idea that somatic cells from the same

Abstract
Inheritable epigenetic modifications, like histone modifications and DNA methylation, were once considered somatically stable and tissue-specific, yet accumulating evidence suggests the contrary.Environmental encounters are transduced into the cell through signaling pathways and these signals are relayed to the nucleus and memorized as epigenetic marks on target genes.Signalingspecific epigenomic changes provide selectable outlines for further lineage determination during differentiation and serve as traceable biomarkers of specific signaling pathways during transformation.Current "big data" interrogation employs signal-specific reduction methods to identify epigenetically modified genes as possible signal targets.These pathway-based analyses show promise for finding true tumor biomarkers, but fail to explain why the onset of cancer can still be partially due to "bad luck".Recent evidence indicates that epigenetically-governed repetitive sequences, especially long and short interspersed nuclear elements (LINE and SINE), in the somatic genome are not evenly distributed.Cell type-specific and even unique single-cell LINE1 transpositions are found in specific brain regions, demonstrating the unevenness of LINE1 in the somatic genome.Retrotransposable elements like LINE1 are silenced, an epigenetic mechanism that stabilizes the genome.Unchained LINE retrotranspositions are found in pre-malignant colon and gastric cancers, and accumulated random transpositions found during the course of cancer development might account for the onset of disease.Epigenetic codes relay and translate cellular encounters into selectable biomarkers and co-evolved with tumorigenesis.Epigenetic regulators are responsible for the maintenance of genomic stability and for the prevention of random transposition.Therefore, changes in epigenetic regulation might explain the probability of oncogenesis and could serve as predictive biomarkers.

ISSN: 2469-5807
Shou-Tung et al.Int J Pathol Clin Res 2016, 2:032 of repetitive sequences [34].These repetitive sequences may be the evolutionary source of variation.For example, the endogenous retrotransposon Long Interspersed Nuclear Element 1 (LINE1) comprises around 17% of the human genome [35].Some LINE1 sites are able to transpose autonomously and can induce harmful disruptions in the genome [36][37][38].Ancient LINE1s are silenced by a distinct set of epigenetic marks; more recently derived LINE1s are mostly silenced by DNA methylation, suggesting that our genome evolved unevenly [39].
It is clear that somatically inheritable epigenetic modifications must fulfill at least two missions during normal differentiation: one is to generate gene expression differences that restrict cell fate changes; the other is to maintain genomic stability during differentiation.If either of these missions fails, a cell might be transformed.Thus, an epigenomic mechanism must be responsible for the changes in gene expression and stabilization or destabilization of the genome during normal somatic stem cell differentiation or tumorigenesis.

Genome Instability might contribute to Transformation
External signals not only may lead to differentiation or transformation, but also might be responsible for the maintenance of genome stability before and after signal-induced changes in cell physiology [2].Repetitive sequences like Short or Long Interspersed Nuclear Elements (SINE or LINE, respectively) are the driving forces behind genomic recombination and evolution [34].Advances in sequencing technology and single-cell genome sequencing have shown that somatic neurons from the same hippocampus region possess different LINE1 insertions (Figure 2).Surprisingly, these variable insertions and transpositions are needed for normal brain function [2,40,41].The stability of these repetitive elements is maintained by epigenetic modifications like DNA methylation and histone modification [42,43].External signals thus induce epigenetic modifications that are needed both to change gene expression and to maintain genome stability before and after differentiation.
There are two proposed molecular mechanisms by which abnormal DNA methylation could lead to cancer [44].One is increased DNA methylation within tumor suppressor genes like HIC1 and RassF1A [15].The other is loss of global DNA methylation due to low methyltransferase activity that might lead to genome instability and ultimately cancer [45].The loss of global methylation is hypothesized to mobilize endogenous transposable elements resulting in random insertions, translocations, and transvections [46].Random transposition could be the "bad luck" element in the onset of cancer [4,41,46].Indeed, random transposition is documented in gastric and colon cancers [39,47,48].Targeted sequencing of LINE1 elements shows LINE1 transposition occurring in pre-malignant tissues and progressively accumulating over the four stages of colon and gastric cancers (Figure 3) [39].The association between LINE1 transposition and the onset and progression of cancer suggests that LINE1 transposition might be one cause of transformation, and furthermore, random transposition might correlate with random onset of cancer [49][50][51].

Epigenomic Biomarkers: Signal-Specific Epigenetic Modifications
Environmental signals like exercise, diet, and smoking are transduced into the cell through more than 131 signaling pathways, including Notch, Wnt, BMP, and estrogen receptor (ER) pathways (Figure 1) [52,53].Changes in a cell's environment can lead to changes in the expression of downstream effector genes and their signal relays [54][55][56][57].Such sequential changes need to be translated into molecular codes via epigenetic modifications to become somatically inheritable [48,58].Accumulated epigenetic changes then result in varied gene expression and establish selectable parameters for further The discovery of signal-induced specific epigenetic changes supports the notion that there is internalization of external signals as  epigenetic modifications [56,62,63].For example, if the expression of ER is knocked down, the YY1-governed repression complex initially silences downstream ER targets [56].Later, DNA methyltransferase (DNMT) adds methyl groups at target loci, and the subsequent passage of cells inherits the methylation pattern [56].This type of abnormal methylation of the ER target loci Trip10, ENSA, and Casp8AP2 takes place in ER non-expressing breast cancers, proving that the same signal-specific epigenetic modifications occur in vivo [56,[64][65][66].
The assembly of co-regulator complexes has made predicting the downstream epigenome even more complicated.To continue with the same ER example, although ER itself is a nuclear receptor, it teams up with MYC, SMAD, and other transcription factors to up-or down-regulate ER target loci.In cis, MYC and SMAD transcription factor binding sites are located around the ER response element and different bound combinations of these transcription factors produce different patterns of gene expression [60,67,68].Therefore, cis-regulatory elements and the architecture of the genome are parts of the machinery that programs cell fates [65,69].Genomic architecture at signaling pathway-regulated sites can make the relay and memorization of signals more efficient.Increasingly, there are blueprints of signal-to-epigenome relays in specific cell types.The results of signals are traceable and the specific patterns of epigenomic changes can reflect the environment around the sequenced cell.Recent research using "big data" has realized the power of reduction studies of the signaling-to-epigenome relay as a blueprint to identify a disease's target genes [70] that may serve as biomarkers.

The Epigenome and Genomic Stability
The three-dimensional (3D) architecture of the genome is critical for genome stability and proper cell physiology [56,[71][72][73].The genome is organized into regulatory units consisting of multiple loops and inter-and intra-chromosomal interactions [74][75][76].With advances in 3C (chromatin conformation capture) and Hi-C techniques [75,77,78], genome-wide maps describing the distribution of these interactions are becoming available.In Caenorhabditis elegans, a recent report indicates that knockout of the genes responsible for sex chromosome dosage compensation can systematically affect genomic interactions within X-chromosome [79].In contrast, the same mutations do not affect the architecture of autosomes, but still interfere with the expression of autosomal genes [79].This finding suggests that 3-D conformation, looping, and interactions among genomic regions are specific regulators of gene expression.Nonetheless, how the compartmentalization of regulatory zones is maintained and how compartments can be changed during differentiation or transformation remains unknown.
Understanding how the epigenome, including the methylome, is established, might help to answer these questions.Mammalian cells are separated into germ line and somatic lineages based on their DNA contents.During the early stages of the development, the methylome of these two kinds of cells undergoes drastic revision [80,81].This methylation process is PIWI-interacting RNA (piRNA)-dependent and may be initiated on repetitive segments within the genome [82].piRNAs are the largest class of small non-coding RNAs in animal cells, and are clustered within the mammalian genome.Argonaute (Ago) proteins can process their transcripts into pieces of RNAs that are further used to cleave target RNAs into smaller pieces [83,84].The resulting smaller RNAs in turn process the remaining piRNA cluster transcripts and further fragment the target RNAs [85][86][87][88].This repeated process is named the piRNA "ping-pong" cycle and it is important to amplify the effects of piRNA activity [89].
piRNAs are also crucial for methylation of their target transcripts' promoters [90].Since piRNAs must enter the nucleus to cause methylation, other factors are needed to direct their translocation and subsequent methylome establishment [91,92].Although the sequences of piRNAs are not conserved across species, piRNAmediated silencing targets might be conserved.Indeed, the function of the Arg genes is conserved across species [93,94].There are eight Arg genes in human cells [95,96].Recently, researchers observed that Arg proteins need an average of only six nucleotides (nts) to recognize their target RNAs [97].During the ping-pong amplification cycle, the frame shift between cycle rounds is about 10 nts.Therefore, it is unsurprising that 6-10 nts are needed for Arg proteins to recognize their piRNA interactors.A similar size recognition sequence may be sufficient for piRNA-mediated methylation silencing of targets such as LINE1 retrotransposons [98,99].In the case of LINE1, 6-10nts is more than enough sequence for target recognition [100,101].However, the impact of subtle sequence differences between different LINE1 elements, located in different genomic regions, is substantial.Even with only two nucleotide differences, LINE1 regions are sometimes silenced by different epigenetic mechanisms [99,102,103].Only the most recently evolved LINE1 copies are silenced by DNA methylation; more ancient copies are not [39].These data suggest that other factors or mechanisms determine the specificity of Arg/piRNA targeting.Some researchers posit that piRNA clusters are the sequences that capture invading alien DNAs and serve as reservoirs of foreign DNAs information [104,105].When transcripts with similar sequences are upregulated, piRNA-targeting sequences are able to silence related endogenous retrotransposons or exogenous retrotransposon-like elements [104].Several constructs with reporter genes have been developed to mimic the capturing system and some of them seemed to be promising until a recent report proved otherwise [106,107].Using a reporter capturing system, the authors disrupted a piRNA cluster sequence by CRISPR/Cas9 and found that the piRNA targeting system was still functional [108,109].Since the investigated piRNA targeting system was piRNA cluster-independent, the search for other factors that are involved in the targeting process is necessary to reveal the links between the methylome and genome stability.

Epigenomic Explanations for Uncertainty in Cancer Predictions Using Biomarkers
The relay of the external signals into cells and the memorization of signals as epigenetic marks on target genes are programmed during differentiation and development.No matter how differentiation progresses, genome stability must be maintained to prevent cellular transformation.Thus, the relay of signals and the induction of epigenetic changes are specific, necessary, and under tight control.However, somatic cells are more mosaic than what was previously appreciated.This is evidenced by the extreme example of hippocampal neurons in which LINE1 transpositions occur within individual neurons (Figure 2) [110].The data suggest that mosaicism is necessary for normal neuronal function [110].Therefore, there must be a fine line that separates good mosaicism and transposition from deleterious changes.It seems that the pivotal separating point may occur during the relay of the external signals into regulatory epigenetic codes.In normal somatic stem cells the relay system might lead to necessary epigenetic modifications but also to possible randomized mosaicism that decreases genomic stability.Further elucidating the molecular explanations for apparently random epigenomic occurrences will assist researchers in findings biomarkers and other epigenetic identifiers associated with disease onset.

Figure 1 :
Figure 1: Signal-governed epigenomic changes.Environmental cues can be transduced into cells, relayed into nuclei, and memorized as epigenetic marks.The unknown mechanism by which external signals are translated into epigenetic codes is marked by a red question mark.

Figure 3 :
Figure 3: Accumulated LINE1 retrotransposition is associated with development of cancers.There are detectable LINE1 retrotransposition in precancer lesions.More and more transposition is observed during tumorigenesis[39].Increased LINE1 transpositions are represented by the increased colored lines in Circos plot.