Highlights
The conference kicked off on a windy September day in Dublin. Autumn had truly fallen on the Epigenetics Europe conference. I woke up early, rolled up my poster, threw on an extra pair of thermals and headed off to the venue on the southern outskirt of the city. Arriving at the large hotel, a friendly American tourist told me that “all the folks with posters went that-a-way”, pointing me in the direction of a stream of eager-looking scientists. I was delighted to talk to three other grad students who told me they had all heard about the conference through EpiGenie.
Four simultaneous sessions were held over the two days. Besides Epigenetics Europe, there was mRNA & RNAi Europe, qPCR Europe and Genomics Automation Europe…did I mention this took place in Europe? The epigenetics hall was over-flowing with people standing outside, straining to get a look-in at Manel Esteller, who kicked the proceedings off with insightful keynotes on the disruption of non-coding RNAs in the epigenetics and genetics of cancer. I caught up with a few of the speakers and asked if they’d mind sharing their work with the EpiGenie community.
Hypoxia Alters DNA Methylation Profiles In Cardiac Fibroblasts
Chris Watson, University College Dublin
Chris gave a fascinating round-up of the research he and the team at UCD have carried out on cardiac fibrosis in response to a hypoxic environment, and how this can be intrinsically related to inflammatory and epigenetic alterations. Chris explained how cardiac fibrosis comes about as a response to tissue injury within the heart, such as ischemic events like myocardial infarction or the inflammation-associated increase in oxygen consumption, which create the hypoxic environment. Hypoxia correlates with increased collagen and inflammatory markers in the cardiac tissue. In Chris’ work, he exposed fibroblasts to 1% O2 using a hypoxia chamber for 4 and 8 days, as well as exposing the cells to TGFß. Thy-1, a membrane-bound glycoprotein, was seen to be hypermethylated in the hypoxic environment. In fact, this low-level oxygen environment induced global alterations in DNA methylation, potentially as a consequence of increased DNMT3b expression. Although the mechanism by which hypoxia regulates DNMT3b expression is not fully characterised Chris’ data suggest an association in fibroblasts with elevated amounts of IL-6 and down-regulation of miRNA 148a. Overall, these findings shed some interesting light on the epigenetic landscape associated with cardiac fibrosis.
Modelling Complex Epigenetic Changes
Dimitri Perrin, Dublin City University and Osaka University, Japan
Dimitri, coming from the field of computational biology, added a refreshing viewpoint to the day’s discussions. He reminded us how it is crucial to understand all epigenetic enzymatic interactions i.e. those between histones as well as DNA methylation, and to avoid succumbing to the tunnel vision so often associated with one’s own extremely specialised research area. Dimitri and his colleagues are developing an exciting framework that will offer epigenetic researchers a host of tools to aid their work. He joked that we can be afraid of computational analysis, but really, it’s there to help us; no matter how scary it seems! At DCU, the team have used a computational model of aberrant DNA methylation induced by H. Pylori, bacteria that colonise the stomach and is a significant factor in development of gastric cancer. Among the new gadgets we can look forward to using are systems to analyse formulated hypotheses, improve the granularity of our results, investigate the influence of multi-faceted parameters in experimental models, assess the probability of aberrant methylation and they’re just the starting points. One of the ideas he mentioned that would, perhaps, be most interesting is a results verification package; something that could save us all hours of manual work. All in all, Dimitri feels confident he can develop the framework for a very powerful epigenetic tool kit.
Dnmt1…Lost In Translation
Diane Lees-Murdock, University of Ulster
Diane spoke about the work in her lab which is part of the Transcriptional Regulation & Epigenetics Research Group at the University of Ulster. They have uncovered a new mechanism of post-transcriptional control of Dnmt1. She explained how they previously found a cytoplasmic polyadenylation element (CPE) in the 3’UTR of the gene (Lees-Murdock et al. 2004 Genomics, 84, 193-204) and have now shown that this sequence is required for efficient translation. Furthermore when they knockdown the protein which binds to this site (CPEB), translation of Dnmt1 is also reduced. The researchers have shown that this element resides within a block of highly conserved nucleotides containing two further motifs for RNA sequence specific binding proteins, pumilio and musashi. Interestingly, it appears that these three motifs combine to control translation, as when the block is removed, translation is abolished. These binding proteins regulate gene expression by shortening and elongation of the mRNA polyA tail and the researchers have shown that Dnmt1 expression correlates with its polyadenylation kinetics. Diane’s group are working hard to further this research and, she assured us, making great progress. So watch this space for more news from the group….
Epigenetics of Aggression In Killer Bees
Douglas Ruden, Wayne State University
We all know the trend at conferences that sees audience numbers dwindle after lunch on the last day. In this case the title alone of the last talk was irresistible enough to somewhat buck that trend. Yes, we were all there for epigenetics, but the words “killer” and “bees”? Absolute bonus material! Having ventured into the field long before it was ever dubbed “epigenetics”, Prof. Ruden has come to be interested in the topic’s relationship with evolution. First things first: killer bees, despite their exciting name, are not highly dangerous predators. They are said to kill only a few people per year (cue disappointed sigh from any sci-fi fans). They are the more aggressive, Africanised cousins of European honey bees. The soldier bees of this family will hunt down a perpetrator of a hive attack for far greater distances than other bee subspecies. Their sting will kill if the swarm is sufficiently big or if the person under attack is an allergy-afflicted Macaulay Culkin and he “can’t see without his glasses”.
Prof. Ruden’s team at Wayne State have used whole-genome shotgun bisulfite sequencing to analyse DNA methylation patterns in the brains of European and African bees. They located 340 loci with at least a 5-fold increase in DNA methylation in African over European bees, and 86 loci that were five times more methylated in European over African bees. The FMRFamide neuropeptide receptor is 17-fold more methylated in African bees; FMRFamides are involved in aggression in mice and defence behaviour in snails. Kainate glutamate receptors induce aggression in mice and these were found to be 10-fold more methylated in European bees than African. GAT1-deletion in mice causes reduced aggression and this was 10-fold more methylated in African bees. Prof. Ruden emphasised the paradox in increased gene expression associated with elevated methylation levels. It is not correct to assume that transcription will be fully blocked by the presence of 5-mC. It is possible that this methylation could be 5-hydroxymethyl, which would have the opposite effect due to its absent protein-binding capacity; but this is not distinguishable by bisulfite sequencing. The talk was peppered with fascinating additional discussions, including the influence of rapid morphological evolution in pure-bred dogs, in which the prof. explained how the British terrier’s strongly downward sloping nose was a relatively recent development and was induced by the loss of a single alanine from a polyalanine chain. Prof. Ruden’s take-home messages were, principally, the ability of stress to alter the epigenome and the heritability of these alterations…and don’t stress out your neighbourhood bees.
CHD7 Regulates Both Nucleoplasmic and Nucleolar Gene Expression
Peter Scacheri, Case Western Reserve University
Peter’s talk on the functions of CHD7 generated a buzz of interest among the audience. Many were previously unfamiliar with CHARGE syndrome, a genetic condition characterised by a complexity of congenital anomalies. The acronym represents the principle physical symptoms of the disorder: Coloboma, Heart malformations, choanal Atresia, Retardation of growth, Genital anomalies and inner and outer Ear malformation. This syndrome is largely caused by de novo mutations in the gene encoding chromodomain helicase DNA binding protein 7 (CHD7). In Peter’s lab they have undergone intense work to map the distribution of CHD7 on chromatin using the method of ChIP-seq. In addition they have characterizd global gene expression patters in wild type, Chd7+/- and Chd7-/- embryonic stem cell lines derived from mouse models of CHARGE syndrome. Collectively, the data indicate that CHD7 binds to distal gene enhancer elements and functions to regulate tissue-specific gene expression. Although CHD7 can modulate genes in both the positive and negative direction, negative directional CHD7 regulation appears to be the most significant effect of CHD7 binding. In addition, extensive work was done investigating the colocalisation of CHD7 to the nucleolus, where it positively regulates rRNA biosynthesis by associating with transcriptionally active rDNA. He also gave an overview of the genetic syndrome Treacher-Collins, which, passed on in an autosomal dominant fashion, results in severely underdeveloped facial tissues. This syndrome is caused by mutations in the TCOF1 gene, encoding treacle, a nucleolar protein that positively regulates rRNA biogenesis. Peter hypothesizes that the mechanisms underlying Treacher-Collins syndrome and CHARGE syndrome might be related.
Imprinted Regulation of PcG Gene Sfmbt2 by Insertion of miRNA
Susannah Varmuza, University of Toronto
Prof. Varmuza’s main research centres around two aspects of mammalian reproductive biology: spermatogenesis and genomic imprinting. At the conference she focused on the latter, giving us an interesting insight into the frequency of imprinted genes in the proximal region of the murine chromosome 2. The team at University of Toronto have identified a single imprinted gene in the proximal region, which was known to harbour an imprinted gene, but was until now unknown. They showed the Polycomb group (PcG) gene, Sfmbt2, is expressed from the paternal allele in early embryos, and in later extraembrionic tissues. To do this, they crossed M. mus domesticus with M. mus castaneus and assessed the imprint status of Sfmbt2 in allelic expression of the F1 embryos. This gene may be involved in trophoblast development by the embryo. Prof. Varmuza explained that Sfmbt2 may provide an anchor for a new domain since imprinted genes generally reside in clusters regulated by an Imprint Control Region (ICR). The team analysed approximately 20 genes within the 3.9Mb domain to discover that Sfmbt2 must be a single gene locus as it is the only imprinted gene in the domain. Interestingly, Prof. Varmuza found a feature distinctive only to mice and rats (not other mammals) in the presence of a large block of miRNAs at intron 10. In rats the gene is imprinted in extraembrionic tissues but, in bovine tissue, Sfmbt2 is biallelic. Furthermore there is evidence that human and pig versions of the gene are also biallelic. A more distantly related rodent, Peromyscus, or North American deer mouse, was tested and found to express Sfmbt2 from both alleles. The size of the Peromyscus Sfmbt2 intron is consistent with its lacking the block of miRNAs; conclusion supported by preliminary sequencing data. All of this indicates that it is the miRNAs that are driving imprinting at this locus in rats and mice.
**EpiGenie would like to thank Elaine Drummond, who is a PhD student at the UCD Institute of Food & Health, University College Dublin for providing us with this conference coverage.