Where does the H3K9me2 come from? LINE repeats have been reported to act as enhancers for 2-cell genes. We found that EHMT2 bound LINEs which, in mESCs, had high H3K9me2 that extended away from the EHMT2 site. This suggests LINEs can act as silencers by promoting H3K9me2. 10/n

One way EHMT2 might act as a barrier to the 2-cell program is by antagonizing the binding of TFs to 2-cell genes. We C&R'd Dppa4, a key TF that regulates ZGA. Many Dppa4 sites were specific to 2CLCs. These sites were enriched at ECORDs and had high levels of H3K9me2. 9/n

The eternal question: how much of this is directly mediated by EHMT2? We ChIP'd EHMT2 and H3K9me2. ECORDs were marked by H3K9me2 and enriched near EHMT2 binding sites. EHMT2 sites near ECORDs often bound to LINE-1 repeats (we'll come back to this in a second...) 8/n

RNAseq on sorted populations showed that, in addition to increasing their numbers, EHMT2 loss strongly affected 2CLC gene expression. We also found DEGs between normal 2CLC and mESCs were clustered, suggesting it may be part of the normal organization of the 2CLC program. 7/n

2-cell like cells (2CLCs) exist at a small proportion of normal mESC cultures. Could our RNAseq actually reflect a change in the numbers of these cells? Yes! We used a reporter and found that EHMT2 depletion increased the numbers of 2CLCs in culture. 6/n

What are ECORDs, though? ECORDs were enriched for genes active in the early, 2-cell totipotent embryo at the onset of zygotic genome activation (ZGA), which was not true of the DEG-UP that were not clustered (Non-ECORDs). 5/n

Strikingly, however, we noticed that a subset of DEG-UPs were clustered along the linear genome. We call these "ECORDS" : EHMT2 Coordinately Repressed Domains. In some cases, we saw >20 genes up-regulated in a row with only 1-2 static genes interspersed. 4/n

We acutely depleted EHMT2 in mESCs using the dTAG system. EHMT2 promotes H3K9me2, a repressive mark, so, as expected, the majority of differentially-expressed genes (DEGs) were up-regulated upon EHMT2 loss. 3/n

Kaushiki and I are excited to share our preprint about fate restriction in pluripotent cells! We show that EHMT2 counteracts the reactivation of clusters of totipotency genes by nucleating heterochromatin domains at LINE-1 repeats. 1/n

@efapostolou29.bsky.social and Stadtfeld Labs
#DevBio 🧬🖥️