Highlights
With Christmas around the corner attendees of the 1st Basic Epigenetic Mechanisms in Cancer conference in Berlin organised by the European Association for Cancer Research were given an early treat at the start of the festival season (aside from the Glühwein!).
The talks and posters provided fascinating insights into the hottest new data in epigenetics and cancer research. The main themes focused on understanding the role of enhancers and related RNAs in cancer development, the links between epigenetics and genome stability, and discovering new drug targets by understanding the function of epigenetic modifiers.
The scientific scope coupled with some excellent discussion facilitators (speed networking anyone?) made the conference great fun and a resounding success. As is to be expected, there were many exciting talks but here is a summary of my favourites:
Epigenetic regulation by MOF containing complexes
Asifa Akhtar | Max Planck Institute of Immunobiology and Epigenetics, Freiburg, Germany
Asifa Akhtar’s group is interested in X-chromosome regulation, and use Drosphila as a model system which, rather than inactivating an X-chromosome as in female mammals, hyper-transcribe genes on the male X chromosome (dosage compensation). The MSL complex, which includes the histone acetyltransferase MOF, is targeted to the X-chromosome in the fly to control this process. Additionally, they have found that MOF is also part of an evolutionarily conserved Non-Specific Lethal (NSL) complex in mammals as well as Drosophila. This is a master epigenetic regulator and has a seemingly innocuous role in regulating housekeeping genes; however, this complex is often misregulated in cancer and Asifa Akhtar has identified a novel role for NSL. By developing a new method to exclusively isolate mitochondrial DNA, they show that members of the NSL complex (MOF and KANSL) are transcription factors for mitochondrial DNA and knocking these proteins out dramatically reduces respiration rates. This has important implications for myocyte cells rich in mitochondria such as those found in the heart and skeleton and indicates as yet unknown links between nuclear and mitochondrial transcription.
DNA modifications: from mechanisms to genome-wide profiling in cancer
Francois Fuks | University of Brussels, Brussels, Belgium
Francois Fuks split his talk into two parts, the first focusing on his group’s work on using DNA methylation profiles to characterise subtypes of breast cancer. He presented working showing that by integrating different genome-wide analyses, such as RNA expression and DNA methylomes, the classification of subtypes of breast cancer increases from four to ten. Three of the methylome-defined subtypes fitted into previously defined categories, HER2+ve, Basal-like and luminal A; however, the other three were new subtypes. Integrating this genome-wide data provides a more comprehensive classification of different breast cancer subtypes, potentially leading to better definitive treatment. The second part of his talk was outlining exciting data they have from Drosophila on the recently discovered posttranslational modification of RNA, 5-hydroxymethylcytosine. To investigate this they have developed hMeRIP (hydroxymethylcytosine RNA immunoprecipitation) to identify modified transcripts, of which there were many. Furthermore, by knocking out the Tet proteins this signal is reduced. However, it remains to be seen how abundant this modification is in the mammalian system and what relevance this has for transcription and/or long non-coding RNA (lncRNA) function.
Role of ncRNAs in transcription and cancer
Tony Kouzarides | Gurdon Institute, University of Cambridge, Cambridge, UK
Tony Kouzarides’ talk also touched briefly on their work investigating posttranslationally modified RNA and the potential for new drugs targets. They are currently developing inhibitors for BCDIN3D which is an RNA methyltransferase that regulates microRNA processing and is important for breast cancer metastasis. They are also on the lookout for other RNA modifiers as potential drug targets for cancer treatment. The remainder of the talk was discussing work which has attempted to extract the functionally important lncRNAs from the noise and to this end they have identified a new class of lncRNAs. By bioinformatically interrogating RNA-seq data from mammalian cell lines they have identified conserved promoters of lncRNAs that are associated with coding genes, naming them – positionally conserved ncRNAs (pcRNAs). Interestingly, the genes that the pcRNAs associated with are highly enriched for developmental transcription factors, e.g. Hox and Sox genes. Furthermore, the pcRNAs are expressed in a very cell type and tissue specific manner, and are co-expressed with the corresponding coding gene. Functional studies show that the pcRNA regulates the gene and knocking them out inhibits gene expression. Surprisingly, overexpressing the pcRNA does not correspond to an overexpression of the gene. To explain this, Tony Kouzarides presented some interesting hypothetical models they are currently testing which link the function of pcRNAs, enhancers and coding gene expression. These pcRNAs are often misregulated in cancer with possible tumour enabling as well as suppressive roles and the expression of some pcRNAs in primary tumours may predict disease outcome.
Epigenetic alterations in oncogene induced senescence (OIS)
Shelley Berger | University of Pennsylvania, Philadelphia, PA, USA
Shelley Berger’s talk put cancer into the context of the changing biology of an organism as it ages. She discussed the balance between the protective and harmful effects of senescence to the cells: In the short term, senescence is protective against cancer as cells stop dividing; however, this process induces increased inflammation which becomes cancer enabling. She explained how there is profound chromatin reorganisation during senescence, including large-scale domain of H3K4me3- and H3K27me3-enriched ‘mesas’ and H3K27me3-depleted ‘canyons’. She then went on to discuss her recently published data in Nature (doi:10.1038/nature15548) which identifies a new tumour protective role of autophagy which works by degrading nuclear components transported to the cytoplasm. This work showed that autophagy-associated protein LC3/Atg8 is present in the nucleus, it interacts with nuclear laminar protein B1 and binds to lamin-associated domains (LADs) on the chromatin. Upon oncogenic insult (but not starvation) lamin B1 is transported to the cytoplasm and degraded in the lysosome. The presence of lamin B1 in the lysosomes was elegantly demonstrated by using cells which stably express a mCherry-GFP-lamin B1 construct. When lamin B1 is present in the nucleus both flurophores are detectable; however, when lamin B1 is transported to the lysosome, the GFP is quenched by the acidic conditions. Degradation of lamin B1 leads to impaired cell proliferation and DNA repair, a beneficial effect to restrain tumourigenic insults. This new area of research indicates that additional nuclear substrates of autophagy will play important roles in physiological and pathological processes.
Enhancer modulation by Dnmt3a and Dnmt3b regulates epidermal stem cell homeostasis: impact on ageing and cancer
Salvador Aznar Benitah | Institute for Research in Biomedicine (IRB Barcelona), Barcelona, Spain
The balance between self-renewal and differentiation is a crucial process for adult stem cells. Loss of this balance predisposes the tissue to ageing and carcinoma formation. Salvador Aznar Benitah showed data outlining the normal changes to the epigenome and epigenetic modifiers during the progression of embryonic stem cell differentiation to keratocytes. His data demonstrated the fruits of mastering technically challenging ChIP experiments using Dnmt3a and 3b showing that these enzymes have a unique, non-overlapping pattern of genomic binding which change upon differentiation. Dnmt3b is found along the gene body of robustly transcribed genes associated with high levels of DNA methylation. Conversely, Dnmt3a binds to the transcriptional start sites of genes required for the interaction of epidermal stem cells with their stroma, but these genes are highly transcribed and devoid of DNA-methylation. Intriguingly, they also showed that these enzymes are enriched at super enhancers and are required for the production of the associated enhancer RNA (eRNA). The binding of these Dnmts at enhancers is dependent on H3K36me3 and results in 5-hydroxymethylation at these regions. Salvador Aznar Benitah discussed the relevance of this to cancer formation as these DNMTs are often mutated in haematological malignancies.
Closing summary
In summary, this was a successful meeting that gave an overview of the many ways that the epigenome and epigenetic modifiers influence the aberrant processes involved in cancer formation. It also offered insights into how, by broadening our understanding of how basic epigenetic regulation is controlled, we can hope to discover new druggable enzymes and pathways to combat this disease.
** Thanks to Emily Saunderson from Barts Cancer Institute for delivering this great summary.