Highlights
Big thanks to EpiGenie reader David Valle-Garcia for making the trip to Puerto Rico for this meeting. We’re sure it was rough David, but thanks for stepping up!
___________________________________
In mid-April the first conference organized by EpiCypher took place in the beautiful city of San Juan, Puerto Rico. The conference was a very intimate gathering of around 100 epigenetic enthusiasts in a gorgeous venue overlooking the Caribbean sea. The conference hosted a very broad range of topics, from ways to apply CRISPR/Cas9 to epigenetic imaging to the latest epigenetic therapies to treat cancer. The environment of the conference was unique and very refreshing thanks to its diversity: a lot of junior PIs had the opportunity to share their exciting science, as well as several representatives from the industry. Here is a brief summary of some of the most interesting talks:
Controlling enhancer methylation: To methylate or to demethylate, that is the question
Yang Shi | Harvard Medical School
We all know that the epigenome is dynamic. But probably we don’t really envision yet how dynamic. On the keynote session of the conference, Dr. Shi gave a very interesting talk about the newly characterized histone reader RACK7 (aka ZMYND8). RACK7 is a homolog of BS69 (aka ZMYND11), an H3.3-specific reader that recognizes H3.3K36me3. In contrast to its homolog, RACK7 is not specific to H3.3 and binds mainly to enhancer and super-enhancer regions rich in H3K4me1 and H3K27ac marks. Interestingly, it also interacts and co-localizes with the H3K4me2/3 histone demethylase KDM5C.
RACK7 KO cells show a decreased recruitment of KDM5C at enhancers and super-enhancers and an increased accumulation of H3K4me3 on these regions. H3K4me3-containing enhancers became super-active and showed increased levels of p300 recruitment and ehancerRNA generation. Moreover, the target genes of the super-active enhancers showed elevated expression levels. Many of these over-expressed genes are linked with migration-related functions and the RACK7 KO cells show a higher migratory and invasion capacity than the wild type counterpart. Thus, it is not surprising that RACK7 is frequently mutated in different cancer types, particularly in breast cancer.
RACK7 is only one example of the dynamics of the epigenome. While we tend to view H3K4me1 as a relatively static enhancer mark, Shi’s lab has demonstrated that it is frequently di and tri-methylated and there is an active mechanism to keep it on its mono-methylated form. It would be interesting to determine if H3K4me3 super-active enhancers (or mega-enhancers? Dr. Shi hasn’t come up with a name yet) are only found in pathological stances or can be a novel type of regulatory region in normal cells.
A combination therapy to fight against pancreatic cancer
Julien Sage | Stanford University School of Medicine
Pancreatic cancer, despite all the efforts of countless scientists, remains one of the least tractable cancers. Despite it’s mainly driven by mutations of the well-known K-Ras oncogene, treatment with Ras inhibitors has been shown largely ineffective. Given the lack of current feasible treatments, the Sage lab devised that epigenetic therapy could be successful there where regular chemotherapy has failed. Using a mouse model for pancreatic cancer, Sage’s lab discovered that treatment with the BET inhibitor JQ1 moderately improved the mice survival and decreased tumor size. These effects, as has been shown for other models, seem to be driven by the inhibition of MYC and other inflammatory signals. Unfortunately, the level of efficacy of the JQ1 treatment doesn’t seem to be huge enough as to make a splash in the clinical world.
Therefore, Sage’s lab tried a combination therapy of JQ1 and different epigenetic drugs known to target different components of the Ras pathway. They found that the combination of JQ1 and the histone deacetylase inhibitor SAHA (aka vorinostat) has a synergistic and very powerful apoptotic effect on pancreatic cancer cells. In order to understand the molecular mechanism that drives this effect, they performed transcriptional profile analysis of JQ1-SAHA treated tumors and found that p57 (aka Cdkn1c) is one of the strongest de-repressed genes upon treatment. Using a novel in vivo CRISPR/Cas9-based methodology, Sage disrupted p57 in the liver of live mice.
Knocking down p57 decreased the apoptotic effect of the JQ1-SAHA therapy and partially reduced its efficacy. Because SAHA is a FDA-approved drug and JQ1 is currently in human trials, this combination therapy is likely to be on the fast track to clinical testing. Good news for those fitting against cancer and a nice example of the potential of epigenetic therapies.
It’s all about the control: A new ChIP-seq methodology for calibrating and comparing datasets
Alex Ruthenburg | University of Chicago
ChIP-seq is the quintessential technique of epigenetics. Thanks to the ever-lowing prices of sequencing it is an affordable and widely used technique to interrogate the epigenome. Unfortunately, standard ChIP-seq is also deeply flawed when it comes to controls for its normalization. For that reason, the Ruthenburg lab developed a novel technique called ICeChIP. ICeChIP stands for Internal Standard Calibrated ChIP. It stems from the simple but bright idea of using spike-ins of barcoded nucleosomes with the histone mark of interest. The internal control allows estimating the efficiency of the pull-down, as well as calibrating the ChIP data in order to make meaningful comparisons between experiments.
Since antibodies are not always entirely specific, adding spike-ins with additional modifications also permits to calculate the rate of false positive and unspecific binding sites. While spike-ins are a relatively standard and widely used method for RNA-seq (or should be) and a very effective way of normalization, it makes sense to import the idea to ChIP-seq that suffers from the same sequencing biases. Hopefully, ICeChIP will soon become the golden standard for ChIP-seq providing us with a meaningful tool to compare across experiments performed in different labs, samples and tissues.
Who reads the readers? Is there a code above the epigenetic code?
Sriharsa Pradhan | NEB
Since its proposal, a lot of evidence has shown the importance of the epigenetic code. However, is there a code over the code? Some sort of meta-epigenetic code? The research from the Pradhan’s group seems to point in that direction. Pradhan has been interested for a good number of years in understanding the function of the DNA methyltransferase DNMT1. Previous research from his group identified two PTMs in DNMT1, the phosphorylation of serine 143 (DNMT1pSer143) and the methylation of lysine 142 (DNMT1K142me1). While DNMT1pSer143 stabilizes DNMT1 during the replication process, DNMT1K142me1 targets it for degradation.
Recently, Pradhan’s group found a reader of DNMT1K142me1: PHF20L1. PHF20L1 is able to recognize DNMT1K142me1 and sequester DNMT1 to pericentric heterochromatin, avoiding its degradation and promoting the DNA mutilation of ribosomal repeats. Furthermore, they now found a reader of DNMT1pSer143: 14-3-3. 14-3-3 binds DNMT1pSer143 post-DNA replication and inhibits its methylation activity. Overexpression of 14-3-3 results in hypomethylation of migratory and proliferation-related genes. Interestingly, human breast cancers that over-express 14-3-3 show also a decreased DNA methylation and reduced survival. It seems then, that there is another code of PTMs regulating the epigenetic regulators, which also has a profound effect in the epigenetic pattern. This code is just being discovered, but it promises to be an interesting layer of regulation that may be exploited in the future for therapeutic proposes.
Summary
There were a lot of exciting talks and unfortunately there is not enough space to talk about all of them. But in general there were at least two big trends that were appreciated. On one hand, young and established PIs are intensively using the CRISPR/Cas9 system to perform genetic manipulations that used to be very time-consuming. Since simple gene disruptions, to more complex gene translocations, to using CRISP/Cas9 as a way to perform live-cell imaging, the CRISPR-based techniques are already revolutionizing the epigenetics field. On the other hand, there is an intense interest in using epigenetic drugs to treat different diseases, particularly cancer.
Since the now well-established BET inhibitors to the more novel EZH2 inhibitors, there are plenty human trials in different stages that are looking promising. There is also a very intense basic research to try to find novel and better inhibitors of epigenetic readers. All in all it was a very exciting meeting with a long list of very promising young PIs and equally interesting research being done by the ever-growing epigenetic industry. I’m already looking forward to the next EpiCypher meeting.