Are epigenetic modifications the most stable mechanisms for cell differentiation?

Question

Wondering what the general take is on what are the molecular mechanisms that are mostly responsible for cell type differentiation stability; ie, for a cell's identity to actually become 'locked in' and progressively less susceptible to extracellular cues (for example, in heterotopic transplantation experiments such as Sue McConnell's).

I'm pretty convinced it really comes down to DNA methylation / chromatin remodelling / histone modifications in order to reach those truly stable, self-perpetuating, differentiated cell identities, but haven't been able to find a paper or review that actually discusses this. Your suggestions and opinions are welcome.

Also posted on reddit/labrats

Edit: I suppose another way to ask this is: Are Waddington's stable "attractor cell states" most likely or most easily explained by epigenetic modifications such as DNA methylation?

Answer 1

Epigenetic marks are reversible (you might be aware of induced pluripotent cells). Many animals can regenerate organs with high tissue complexity (such as a limb) and this involves de-differentiation in some species (Sandoval-Guzman et al., 2014). Even otherwise, cells can respond to extracellular/environmental cues to modify their epigenetic state (for e.g. EMT involved in wound healing). I would still consider epigenetic marks to be more "stable" than simple regulatory mechanisms.

Only way for a cell's identity to actually become 'locked in' is by genetic changes. The maturation of lymphocytes (especially the B-cells), would be a good example. These genetic changes happen via V(D)J recombination, class-switching recombination and somatic hypermutation.

L1 retrotransposons are also known to contribute to somatic heterogeneity in the brain neurons. However, the frequency of insertions is not very high (Evrony et al., 2012).