Fig. 2

The MV core histone locus defines a full repertoire of basal eukaryotic core histones. a All MV genomes possess a pair of divergently transcribed histone doublet genes encoding a predicted protein with fused H2B and H2A moieties (H2B-H2A) and another predicted protein with fused H4 and H3 moieties (H4-H3). This gene pair thus encodes obligate H2B-H2A and H4-H3 dimers. b Bayesian MCMC-based inference shows that each MV core histone domain defines a well-supported sister clade (yellow highlighted clades with red lineages) to the eukaryotic core histone groups, including eukaryotic core variants for H2A.Z/H2A and H3/CenH3 (purple and blue clades within each core family). Posterior probabilities following 2,000,000 generations of sampling mixed amino acid substitution models with 25% burn-in generations and without Metropolis coupling (heated chains) are indicated at all nodes. The average standard deviation of split frequencies from two parallel runs was less than 1% (0.0076). This analysis indicated posterior probabilities of 80.5 and 19.5% for the Wag [58] and Blosum [59] amino acid models, respectively. All core histone clades from eukaryotes and MV viruses are grouped in a single super-clade of the core histones with a posterior probability of 0.99. Representative archaeal histones are grouped together in an out-group clade at the bottom. This analysis was conducted on an alignment using MUltiple Sequence Comparison by Log-Expectation (MUSCLE) [32]. c The above phylogenetic analysis suggests that that the basal core histones predate eukaryote-specific duplications and neofunctionalizations in the Hα and Hγ clades. Interestingly, some of these eukaryote-specific specializations are associated with intergenic nucleosomes (eukaryotic canonical H2As) or centromeric nucleosomes (cenH3s). Thus, the core basal histones, defined as Hα, Hβ, Hγ, and Hδ likely predate the evolutionary innovation of large, linearized chromosomes with centromeric pairing mechanisms, a late stem-eukaryotic innovation