Ctenophore methylation and chromatin looping

Today the paper led by our collaborators Iana Kim and Arnau Sebe-Pedrós lab in Barcelona got published in Nature: https://www.nature.com/articles/s41586-025-08960-w

Unsurprisingly, our contribution to this work, which deals with the origins of chromatin looping in animals, lies in DNA methylation. In a previous study, we sequenced the methylome of the ctenophore Mnemiopsis leidyi, which was quite puzzling. Unlike other invertebrates, gene bodies are not particularly enriched for 5mCG, and promoters are not unmethylated, which is typical of H3K4me3 marked regions. Instead, genes could have methylated "promoters" (proximal upstream regions to the start position of the gene), which were characterised with high repeat/Transposable Element content. Having methylated repeats in the promoter didn't seem to affect gene expression, as there were genes at all levels of expression that had these weird promoters.

Adapted from de Mendoza et al 2020, Nat Ecol Evol

At some point, I thought this was just an artifact of poor gene prediction, as perhaps those "promoters" were not really promoters but truncated gene models, but back then we didn't have the data to proof that....

Some years later, when Arnau's group obtained a new chromosome scale version of the Mnemiopsis leidyi genome, we improved gene annotation with a comprehensive approach and new methods, and checked methylation again. Still, 5mCG was found on those repeat-rich "promoter" regions. So how does that work? Well, it seems Ctenophores have a lot of chromatin loops, which put these genes with repeat-rich methylated promoters in contact with H3K4me3 marked distal "enhancers", which perhaps are playing the role of "promoters" in this species.

So how did ctenophores end up with this weird genomic configuration? It could be that methylation was targeted to repeats in general, also those colonising promoters, silencing them. This accumulation of junk on a sensitive regulatory region was then tolerated by looping over them. It could also be that the looping requires methylation, as the structural CTCF-like zinc fingers that mediate these loops seem to be methyl-sensitive. In sum, ctenophores represent a very interesting example of animals with methylation restricted to transposable elements, which show further links to 3D chromatin structure and genome regulation.