Over the last two decades it has become clear that epigenetic modifications acquired by an individual during its lifetime can be inherited for multiple generations. There are a growing number of examples where a clear case can be made for the inheritance from parent to offspring of environmentally acquired gene expression changes. We have developed a transgenerational epigenetic inheritance (TEI) sensor in the model organism Caenorhabditis elegans in which RNAi-induced silencing of a GFP transgene is robustly inherited for multiple generations
We have found that two putative histone methyltransferases, SET-9 and SET-26, are involved in establishment and long-term maintenance of silencing, respectively. Intriguingly, the methyltransferase domains of these two proteins are not well conserved and their ability to act as histone lysine methyltransferases is controversial. SET-9/26 also contain PHD finger domains that bind H3K4me3 with high affinity in vitro. TEI assays in strains containing inactive methyltransferase or PHD finger domains show that the methyltransferase domain is required for establishment, whilst the PHD finger’s role is mainly in maintenance.
Interestingly, we also show that a catalytically inactive version of the H3K9 methyltransferase SET-25 is only partially defective in TEI, implicating other protein domains in SET-25’s role in TEI. We have identified a previously uncharacterised domain in SET-25 that is similar to a chromodomain, and which is also required for TEI.
Taken together these data suggest a model whereby SET-25, SET-9 and SET-26 bind methylated lysines using their chromodomain and PHD finger domains, and methylate neighbouring histones to nucleate and spread the silencing signal. Effective establishment and transmission of the epigenetic mark between generations requires both methyltransferase activity and methyl reader activity.