All tags Epigenetics Top Epigenetics articles: May 2014

Top Epigenetics articles: May 2014

Bombarded by TOC alerts? Want to know who’s publishing on the cutting edge this month? Here are Abcam’s must read papers for May.

Inhibition of intestinal tumor formation by deletion of the DNA methyltransferase 3a.

Dnmt3a deletion stalls intestinal tumors.

DNA methyltransferase 3a (Dnmt3a) is one of three members of the Dnmt3 family. It mediates DNA methylation and is a required element in development. However, Dnmt3a is also responsible for the de novo catalysis of DNA methylation in mammalian cells at CpG islands, which may sometimes lead to genomic instability and aberrant gene silencing.

Dnmt3a has been suspected of playing a part in a number of cancers, so to investigate Dnmt3a’s de novo role in intestinal tumors, researchers from the German Cancer Research Center (DKFZ) in Heidelberg, Germany (Oncogene, May 2014) studied Dnmt3a expression in intestinal tumor formation with a variety of profiling techniques including RNA fluorescence in situ hybridization (FISH), quantitative PCR and immunostaining on colon crypts, murine colon adenomas and human colorectal cancer samples.

To better understand the precise role of Dnmt3a, the group conducted loss of function experiments by conditionally deleting Dnmt3a in APC(Min/+) mouse colons, then assessing the tumor number, genotypes, DNA methylation and gene expression. Here’s what they learned:

  • APC(Min/+) colon adenomas had increased Dnmt3a expression.
  • Dnmt3a deletion inhibited tumor formation by about 40%.
  • The vast majority of remaining colon tumors still contained the functional Dnmt3a allele, while none showed inactivated Dnmt3a.
  • Dnmt3a deletion led to regional DNA methylation loss.
  • Deletion of Dnmt3a caused promoter demethylation (at Oct4, Nanog, Tff2 and Cdkn1c promoters) and increased tumor-suppressor gene expression (including Tff2 and Cdkn1c).

The authors conclude that Dnmt3a is predominantly found in tumor stem and progenitor cells and that removal of Dnmt3a stalls tumor growth at a very early stage.

Find all of the Dnmt3a details at Oncogene, May 2014.



The tissue-specific transcriptomic landscape of the mid-gestational mouse embryo.

First complete tissue-specific transcriptomes of the mouse embryo.

Understanding differential gene expression and the gene regulatory networks that control the differentiation of various cell lineages requires a thorough survey of the complete transcriptomes of individual cell and tissue types. In a recent publication, (Development, May 2014) scientists from the Max-Planck-Institute for Molecular Genetics unveiled the first complete transcriptome maps from mouse embryos.

This data set is invaluable for future expression profiling research as the first to represent the mouse embryonic transcriptome. In addition, it addresses other areas that had yet to be explored by earlier projects including complete coverage of protein coding genes, alternative transcripts and noncoding RNA genes.

The team applied an RNA-seq approach to gather details about the transcriptomes of six tissue types harvested from the embryos of TS12 mice.

After sorting through nearly one billion reads, here is what the researchers accomplished:

  • RNA deep sequencing revealed 1375 genes with tissue-specific expression, generating gene signatures for all six TS12 tissues.
  • Cataloged alternative splicing and extended UTRs.
  • Annotated 1403 novel long noncoding (lncRNA) transcripts and found 439 that had tissue-specific, differential expression.
  • Discovered 392 divergent coding-noncoding gene pairs (CNPs) that likely share promoters.
  • Confirmed that H3K4me3 marks act to support alternative TSS and novel transcripts.

The study outlines the first complete transcriptome map compiled for the mouse embryo, and provides a foundation and data resource for further investigation into the genes and genomic networks that guide mid-gestational development.

Find the complete  transcriptome analysis in Development, May 2014.



PADI4 acts as a coactivator of Tal1 by counteracting repressive histone arginine methylation.

Histone arginine methylation regulates transcription factor Tal1.

Tal1 is a transcription factor that is important to the modulation of gene expression in hematopoiesis and leukemia, however its mechanism of influence on specific genes is still unclear.

Tal1 can interact with histone modifying proteins. Depending on which of these Tal1 interacts with, dictates whether Tal1 represses or activates gene expression.

Also, since histone-modifying enzymes can be targeted with therapeutic approaches, identifying Tal1s interactions, would not only provide insights into its gene regulation mechanisms, but also identify novel therapeutic targets.

A recently published report (Nature Communications, May 2014) uncovered new details about the role of Tal1.

A team led by Jörn Lausen at the Georg-Speyer-Haus, Institute for Tumor Biology and Experimental Therapy employed a wide range of techniques including SILAC, ChIP, shRNA Knockdown and Gene Expression Microarrays to better understand the Tal1 mechanism.

Here is what the researchers learned:

  • Peptidylarginine deiminase IV (PADI4) is a novel Tal1 interaction partner.
  • PADI4 and Tal1 act in concert to increase gene expression of IL6ST.
  • PADI4 acts as an epigenetic co-activator by influencing histone arginine methylation (H3R2me2a)

The interleukin 6 signal transducer (IL6ST) is an intermediate of the interleukin-6 signaling pathway and a target of Tal1. The research team found that repressive H3R2me2a marks are counteracted by PADI4 at the IL6ST locus, thereby reinforcing active H3K4me3 marks and generating elevated IL6ST expression.

The authors propose that PADI4’s control of IL6ST transcription manipulates IL6ST expression during the differentiation of hematopoietic stem and progenitor cells.

This model suggests the future possibility of pharmacological treatments that target Tal1 in leukemia therapies.

Get the latest on Tal1 at Nature Communications, May 2014.



RLIM is dispensable for X-chromosome inactivation in the mouse embryonic epiblast.

Xist readied for mouse X-chromosome inactivation without RLIM.

The long non-coding RNA Xist plays a key role in X-chromosome inactivation (XCI) in mice, but researchers are still unsure of how Xist is activated at different stages of embryonic development. Previous studies have indicated that the ubiquitin ligase RLIM was responsible for Xist activation, but recent data (Nature, May 2014) suggest this may not always be true.

During development of female mice two forms of XCI occur. Very early on, at the four-cell stage, imprinted XCI (iXCI) takes place where Xist expression is triggered by RLIM. Later in the epiblast, random XCI (rXCI) restores the paternal X chromosome, however it is unclear whether or not RLIM is needed to activate Xist for random XCI.

A research team from University of Massachusetts Medical School punlished the results of their investigation into the role of RLIM in random XCI. Here is what they found:

  • RLIM is required for imprinted XCI in mice, but not for random XCI.
  • Embryonic cells undergoing random XCI displayed lowered RLIM levels.
  • Genetic analysis of female cells devoid of RLIM from pre-implantation stages still exhibited XCI.
  • Xist clouds, H3K27me3 focus, and embryonic potential still remained without the presence of RLIM.

The authors conclude that, based on their evidence, RLIM is unnecessary for random XCI, which hints at the existence of some other, RLIM-independent mechanism in mice that activates Xist during epiblast development.

See the full report in Nature, May 2014.



The noncoding RNA IPW regulates the imprinted DLK1-DIO3 locus in an induced pluripotent stem cell model of Prader-Willi syndrome.

Long Non-coding RNAs Found to Regulate Genomic Imprinting.

Long non-coding RNAs (lncRNAs) were not previously thought to influence imprinting, but a new report reveals that some lncRNAs can facilitate regulation between imprinted genomic regions. This novel mechanism for the regulation of gene expression was reported by scientists at Hebrew University (Israel) who discovered the first known instance of lncRNAs from a paternal imprinted loci regulating the expression of maternally imprinted regions; indicating a cross-talk between those two imprinted regions.

The researchers generated induced pluripotent stem cells (iPSCs) from Prader-Willi Syndrome (PWS) patient samples for use as an investigatory model, and found that that the SNRPN-UBE3A and DLK1-DIO3 imprinted clusters of ncRNA interact via the novel lncRNA IPW. Here are more results from the experiments:

  • Maternally expressed genes from the DLK1-DIO3 locus were found to be significantly upregulated.
  • This expression pattern was unexpected based on the chromosomal location, on chromosome 14, which displays an opposite imprinting pattern, and had not been linked to PWS.
  • The novel lncRNA IPW, which stems from the PWS locus, was demonstrated to regulating the transcripts from the other imprinted cluster.
  • Altered DLK1-DIO3 transcripts correlate more significantly with H3K9 trimethylation chromatin modifications rather than DNA methylation.

The authors hypothesize that certain, defining phenotypes associated with the SNRPN-UBE3A locus may in fact be caused by the interruption of regulatory communication between chromosomes via lncRNA.

See the full report at Nature Genetics, May 2014.



Inhibitors of enhancer of zeste homolog 2 (EZH2) activate tumor-suppressor microRNAs in human cancer cells.

EZH2 Inhibitors Activate Tumor-Supressor miRNAs.

Enhancer of zeste homolog 2 (EZH2) is a known enhancer of tumorigenesis and is overexpressed in various cancers, which has made EZH2 inhibitors important therapeutic targets. A recent study profiled miRNA expression in gastric and liver cancer cells treated with two different EZH2 inhibitors, suberoylanilide hydroxamic acid (SAHA) and 3-deazaneplanocin A (DZNep), to gain a deeper understanding of their anti-cancer properties.

Researchers at Keio University in Japan first proved that SAHA and DZNep did indeed repress EZH2 expression, as well as proliferation in AGS and HepG2 cells. The team then moved on to examine the miRNA profiles in their samples. Here are some of the results:

  • Microarray analyses revealed that miR-1246 is often upregulated in both SAHA and DZNep treated cancer cells, while miR-302a was upregulated by SAHA, and miR-4448 expression was enhanced by DZNep treatments.
  • The genes DYRK1A, CDK2, BMI-1 and Girdin, (targets of miR-1246, miR-302a and miR-4448), were suppressed by SAHA and DZNep exposure, inducing apoptosis, cell cycle arrest and lowered cell migration.
  • ChIP assays showed that SAHA and DZNep reduced binding of EZH2 to miR-1246, miR-302a and miR-4448 promoter regions.

Based on this new evidence, the authors propose that EZH2 inhibitors, including SAHA and DZNep, function as anti-cancer agents by activating the expression of specific tumor-suppressor miRNAs.

See the full report in Oncogenesis, May 2014.



Variant PRC1 Complex-Dependent H2A Ubiquitylation Drives PRC2 Recruitment and Polycomb Domain Formation.

Polycomb Domain Formation Triggered by Histone H2A Ubiquitylation.

The polycomb repressive complexes PRC1 and PRC2 have a well-established role as chromatin modifiers, despite the fact that their precise targeting and functional mechanisms have yet to be deciphered. Researchers at the University of Oxford developed a de novo targeting assay in mouse embryonic stem cells (ESCs) to reveal what makes PRC1 and PRC2 so critical in gene regulation, differentiation, and development.

The scientists and their newly built assay discovered several important and surprising results, including:

  • A PRC1 variant triggers Histone H2A ubiquitylation, H2AK119ub1, causing PRC2 binding and subsequent H3K27 trimethylation.
  • Ubiquitylation is dependent on a variant PRC1 complex, normal PRC1 complexes are unable to deposit H2AK119ub1 or recruit PRC2.
  • Variant KDM2B/PCGF1/PRC1 complex is necessary to enable polycomb domains to form at CpG Islands (CGIs).
  • When KDM2B/PCGF1/PRC1 targeting is absent, it results in aberrant and lethal polycomb phenotypes in mice.

Deposition of PRC1 to target sites was thought to occur through a process that worked through prior nucleation of PRC2 and then addition of H3K27me3. These new findings show an unexpected PRC1-dependent mechanism for PRC2 recruitment to in vivo target sites.

Find the complete article at Cell, May 2014.



Genome-wide analysis reveals characteristics of off-target sites bound by the Cas9 endonuclease.

Genome-wide Analysis of Cas9 Off-target Sites.

With exploding interest in the field of genome editing, and the CRISPR-Cas9 system in particular, new attention is being paid to the details of targeting specificity and off-target cleavage of the Cas9 endonuclease.

Scientists at the University of Virginia took a genome-wide look at Cas9 binding by using chromatin immunoprecipitation and high-throughput sequencing (ChIP-seq) to map the binding sites of an inactive version of Cas9 (dCas9). The study used HEK293T cells and a set of 12 different single guide RNAs (sgRNAs) to investigate where in the genome Cas9 might be directed. Here is what the team found:
•    dCas9 bound off-target sites ranged from around 10 to over 1,000, depending on the sgRNA.
•    The PAM (Protospacer Adjacent Motif)- proximal region of a sgRNA guiding sequence was found to be crucial to target specificity.
•    Stretches of open chromatin were enriched with dCas9 binding sites.
•    With active Cas9, indels at off-target binding sites identified via ChIP-seq were observed, but at significantly lower rates than on-target sites.

This paper provides new insights into the factors influencing Cas9 targeting, and demonstrates that ChIP-seq is a useful technique for the identification of Cas9 binding sites across an entire genome.

See the entire report in Nature Biotechnology, May 2014.



Uncoupling transcription from covalent histone modification.

Histone Modifications Not Always Responsible for Transcriptional Regulation.

The key concept that histone modifications act to regulate RNA transcription levels has been a foundation of chromatin research, but new evidence shows that is not always the case. Researchers at Louisiana State University and Memorial Sloane-Kettering Cancer Center found an exception to this rule by showing that they could induce gene activation in the heterochromatin of yeast without histone modifications.

The team created inducible transgene systems to monitor transcription levels in S. cerevisiae. Here is what their model system revealed:

  • Large, over 200-fold, inductions of heterochromatic transcription were possible.
  • Little histone loss and very low levels of H3K4 trimethylation, H3K36 trimethylation and H3K79 dimethylation were noted.
  • There was minimal H3 and H4 lysine acetylation, and no switch from H2A to the transcription friendly H2A.Z.
  • Absence of histone modification was not a result of reduced transcriptional output.
  • RNA pol II was occupancy was unblocked in activated heterochromatic promoter and coding regions of cells without H3K79 methylase activity.

Taken together the study finds that large increases in transcription can occur despite a virtual absence of histone modifications, which had been thought critical to gene activation. This significant data runs counter to conventional chromatin understanding, suggesting that gene transcription may take place in living cells without any covalent modifications of the chromatin template.

Get the full report at PLoS Genetics, April 2014.



Register to receive our monthly top epigenetics articles.

Sign up