Histones provide one of biology’s most familiar solutions to the problem of organizing DNA. In eukaryotes, DNA wraps around histone proteins to form nucleosomes, creating the basic structural unit of chromatin. But histone-based genome organization is not limited to eukaryotes. Related systems have also been found in archaea and large DNA viruses, raising a basic question: how much can the molecular details of a nucleosome change while preserving the same general architecture?
Melbournevirus provides a particularly unusual example. Its histones are highly divergent from their eukaryotic counterparts and occur as fused histone doublets; yet, they still assemble into a recognizable nucleosome-like particle. The cover image of the September 15 issue of Biophysical Journal is based on Figure 1c of our study and compares the electrostatic surfaces of the Melbournevirus histone core and a canonical eukaryotic nucleosome. Although the two structures look broadly similar, their surface properties are strikingly different. The eukaryotic histone core has a strongly basic DNA binding surface, while the corresponding surface of the Melbournevirus histones is much more neutral.
We generated the image by calculating electrostatic surface potentials with the Adaptive Poisson-Boltzmann Solver and mapping them onto the molecular surfaces in ChimeraX. The DNA is shown separately so that the difference in surface charge can be seen while preserving the structural context of the protein-DNA interface. This contrast reflects what we observed in our molecular dynamics simulations. The Melbournevirus nucleosome-like particle showed more DNA unwrapping, weaker and more transient histone DNA interactions, and greater flexibility than the canonical eukaryotic nucleosome. Thus, two systems can adopt similar structures while relying on quite different physical interactions.
For us, that is the central idea captured by the image. Structural similarity does not necessarily imply similar molecular behavior. By comparing divergent nucleosome systems, we can begin to identify which physical features are essential for organizing DNA and which can vary substantially while preserving the same basic structural solution.
Additional computational biophysics research from the Wereszczynski Research Group can be found at wereszczynskilab.org.
— Melanie E. Melo and Jeff Wereszczynski