Table of Contents

Volume 2, Issue 4

  October/November/December 2007

  Pages 201 - 243

View covers archive

REVIEWS

Open Access Article

What Are Imprinted Genes Doing in the Brain?

William Davies, Anthony R. Isles, Trevor Humby and Lawrence S. Wilkinson
Pages 201 - 206
http://dx.doi.org/10.4161/epi.2.4.5379
Abstract | PDF

As evidence for the existence of brain‑expressed imprinted genes accumulates, we need to address exactly what they are doing in this tissue, especially in terms of organizational themes and the major challenges posed by reconciling imprinted gene action in brain with current evolutionary theories attempting to explain the origin and maintenance of genomic imprinting. We are at the beginning of this endeavor and much work remains to be done but already it is clear that imprinted genes have the potential to influence diverse behavioral processes via multiple brain mechanisms. There are also grounds to believe that imprinting may contribute to risk of mental and neurological disease. As well as being a source of basic information about imprinted genes in the brain (e.g., via the newly established website, www.bgg.cardiff.ac.uk/imprinted_tables/index.html), we have used this chapter to identify and focus on a number of key questions. How are brain‑expressed imprinted genes organized at the molecular and cellular levels? To what extent does imprinted action depend on neurodevelopmental mechanisms? Do imprinted gene effects interact with other epigenetic influences, especially early on in life? Are imprinted effects on adult behaviors adaptive or just epiphenomena? If they are adaptive, what areas of brain function and behavior might be sensitive to imprinted effects? These are big questions and, as shall become apparent, we need much more data, arising from interactions between behavioral neuroscientists, molecular biologists and evolutionary theorists, if we are to begin to answer them.

Open Access Article

Control of Imprinting at the Gnas Cluster

Jo Peters and Christine M. Williamson
Pages 207 - 213
http://dx.doi.org/10.4161/epi.2.4.5380
Abstract | PDF

Genomic imprinting is a form of epigenetic regulation in mammals whereby a small subset of genes is silenced according to parental origin. Early work had indicated regions of the genome that were likely to contain imprinted genes. Distal mouse chromosome 2 is one such region and is associated with devastating but ostensibly opposite phenotypes when exclusively maternally or paternally derived. Misexpression of proteins encoded at the Gnas complex, which is located within the region, can largely account for the imprinting phenotypes. Gnas is a complex locus with extraordinary transcriptional and regulatory complexity. It gives rise to alternatively spliced isoforms that show maternal‑, paternal‑ and biallelic expression as well as a noncoding antisense transcript. The objective of our work at Harwell is to unravel mechanisms controlling the expression of these transcripts. We have performed targeted deletion analysis to test candidate regulatory regions within the Gnas complex and, unlike other imprinted domains, two major control regions have been identified. One controls the imprinted expression of a single transcript and is subsidiary to and must interact with, a principal control region that affects the expression of all transcripts. This principal region contains the promoter for the antisense transcript, expression of which may have a major role in controlling imprinting at the Gnas cluster.




RESEARCH PAPERS

Open Access Article

MeCP2 Deficiency in the Brain Decreases BDNF Levels by REST/CoREST-Mediated Repression and Increases TRKB Production

Liron Abuhatzira, Kirill Makedonski, Yotam Kaufman, Aharon Razin and Ruth Shemer
Pages 214 - 222
http://dx.doi.org/10.4161/epi.2.4.5212
Abstract | PDF

Disruptions in the expression of the Bdnf gene that encodes a neurotrophic factor involved in neuronal survival, differentiation and synaptic plasticity has been proposed to contribute to the molecular pathogenesis of Rett syndrome. Rett syndrome (RTT) is a neurodevelopmental disorder, caused by mutations in the X-linked methyl CpG binding protein 2 gene (MeCP2). MeCP2 deficiency in the brain has been shown to decrease overall expression of Bdnf in spite of an observed increase in the activity of promoter III that appears to be controlled directly by MeCP2. Therefore, how MeCP2 deficiency causes an overall down regulation of Bdnf expression was an enigma. Here we report that MeCP2 deficiency in human and mouse brain causes an increase in expression of two neuronal gene transcriptional repressors REST (RE1 silencing transcription factor), and CoREST. MeCP2 binds to and is involved in repression of Rest and CoRest promoters despite their unmethylated state. MeCP2 depletion is associated with a change in the histone modification profile to a more active conformation. The elevated levels of REST and CoREST in the brain of RTT patients and MeCP2 deficient mice result in down regulation of Bdnf, apparently by their binding to the RE1 (element) located between the first two promoters of the Bdnf gene. Interestingly, the Trk2 gene that encodes the BDNF receptor, TRKB, was overexpressed in MeCP2 deficient human and mouse brains either directly or as an attempt to compensate for BDNF deficiency.

Open Access Article

Global DNA Hypomethylation in Liver Cancer Cases and Controls: A Phase I Preclinical Biomarker Development Study

Rafael Guerrero-Preston, Regina M. Santella, Adolfo Blanco, Manisha Desai, Maria Berdasco and Mario Fraga
Pages 223 - 226
http://dx.doi.org/10.4161/epi.2.4.5214
Abstract | PDF

Background: Global genomic DNA hypomethylation is a feature of genomic DNA derived from solid and hematologic tumors in animal models and human carcinogenesis. Global genomic DNA hypomethylation may be the earliest epigenetic change from a normal to a pre-malignant cell. Objectives: To test if global hypomethylation is a good marker for early detection of cancer we used a novel quantification method of 2’-deoxynucleosides to evaluate DNA methylation in liver cancer cases and controls. Methods: Frozen tissue from liver cancer patients and controls were obtained from the Cooperative Human Tissue Network. DNA was extracted using standard methods. Genomic DNA samples were boiled and treated with nuclease P1 and alkaline phosphatase. Global genomic DNA methylation patterns were obtained using HPLC for fraction separation and mass spectrometry for quantification. A two-sample t test was performed using Welch’s approximation for samples with unequal variances. A Wilcoxon rank sum test was also performed. Results: A global genomic DNA methylation index measuring methylated cytidine relative to global cytidine in the genome was significantly lower (p-value = 0.001) for all cases, mean = 2.43 (95% CI, 2.08, 2.78), when compared to controls, mean = 3.55 (95% CI, 3.16, 3.93). Discussion: A correlation between global genomic DNA methylation patterns and type of liver tissue was observed. These results add to the accumulating body of evidence suggesting that global DNA hypomethylation may be a useful biomarker to distinguish between liver cancer cases and controls.

Open Access Article

Correlation of DNA Methylation with Histone Modifications Across the HNRPA2B1-CBX3 Ubiquitously-Acting Chromatin Open Element (UCOE)

Marianne Lindahl Allen and Michael Antoniou
Pages 227 - 236
http://dx.doi.org/10.4161/epi.2.4.5231
Abstract | PDF

The methylation-free CpG island encompassing the divergently transcribed promoters of the HNRPA2B1 and CBX3 housekeeping genes possesses a dominant ubiquitously-acting chromatin opening element (UCOE) capability. This element allows reproducible and stable transgene expression including from within centromeric heterochromatin. We present an investigation of DNA methylation and histone modification marks across the HNRPA2B1-CBX3 locus in primary peripheral blood mononuclear cells (PBMCs) to characterise the chromatin structure that underlies UCOE activity. The CpG methylation-free region associated with the UCOE extends into the central areas of HNRPA2B1 and CBX3, with a total length of ~5kb. However, the DNA in the 3’ half of both genes is methylated. Histone H4 lysine (K) acetylation shows a broad distribution across both genes, whilst histone H3 lysine acetylation peaks around the transcriptional start sites and drops to background levels at the 3’ ends. Higher levels of H3K4 di-methylation are present at the 3’ end of the genes in contrast to H3K4 tri-methylation which peaks around the transcriptional start sites. Therefore, methylated DNA in transcribed regions of these genes has been shown here to co-exist with active histone modification marks, indicating that these functionally opposing epigenetic signatures can overlap. This suggests that an extended large region of unmethylated DNA in combination with distinct histone modification patterns are at the basis of UCOE function.

Open Access Article

The EPH Receptor Bs (EPHBs) Promoters are Unmethylated in Colon and Ovarian Cancers

Qinghua Wu, Guro E. Lind, Hans-Christian Aasheim, Francesca Micci, Ilvars Silins, Claes G. Tropé, Jahn M. Nesland, Ragnhild A. Lothe and Zhenhe Suo
Pages 237 - 243
http://dx.doi.org/10.4161/epi.2.4.5406
Abstract | PDF

Aberrant expression of EPH receptors and their ligands, ephrins, has been reported in a large variety of human cancers, including epithelial cancers from the colon and ovary. Due to the recently reported decrease or loss of EPHBs expression in colorectal carcinomas and the abundance of CpG sites in their promoters, we analyzed the promoter methylation status of three members of the EPHB family, EPHB2, EPHB3 and EPHB4, in a series of 22 colon cancer cell lines, as well as in four ovarian cancer cell lines and 56 ovarian tumor samples. The promoters of the three receptor genes were unmethylated in the vast majority of samples as assessed by methylation-specific polymerase chain reaction (MSP). These results were confirmed by direct bisulphite sequencing. Furthermore, from RT-PCR analyzes and Northern blotting, EPHB2 showed only small variation in RNA expression across ovarian cancer cell lines and clinical samples. We conclude that promoter hypermethylation of EPHB2, EPHB3 and EPHB4 is not a common event in colon and ovarian cancers, and therefore plays no major role in these tumors.