The TAFs in the HAT

The striking phenomenon of histone acetylases with histone-like substructures that play critical (though distinct) roles in eukaryotic gene regulation raises evolutionary questions. We propose the following speculative scenario that begins with an Archea-like organism containing TBP and histones lacking N-terminal tails. Following the split between Archea and eukaryotes, we imagine a eukaryotic ancestor with histone tails and a histone acetylase that might conceivably be associated with the histones; such an organism would have the ability to modify nucleosome structure. Subsequently, gene duplications and evolutionary divergence results in two sets of histones; the standard nucleosomal histones and the histone-like TAFs that associate with the histone acetylase while losing the ability to form nucleosomes. Next, the primitive histone acetylase complexes diverge into two types, which are distinguished by the catalytic subunit (Gcn5 or TAF-HAT). The TAF-HAT type acquired the ability to interact with TBP, additional TAF subunits, and promoter DNA, whereas the Gcn5 type acquired Ada and Spt subunits to facilitate the interaction with (and hence modification of) nucleosomal histones. Further divergence after the yeast–human split generated novel TAF-like proteins (e.g., PAF65α and PAF65β) that are specific to the human PCAF and hGcn5 complexes. In considering these ideas, it would be of interest to examine very primitive eukaryotes for the presence of histones, TAFs, Spt and Ada proteins, and Gcn5-like histone acetylases.