I guess that H2B is mainly used because it's the most represented canonical histone in nucleosomes during this step of cell cycle. In fact H2A is known to have a lot of variant, has H3. I don't know for H4.
Also I do agree with Vicente & Marceline, H2B-GFP is well-studied and used for many years.
Because the Green Fluorescent Protein (GFP) has been well-studied and characterized over the years, has been proven to work, as well as convenient and easy to use. Perhaps you can discover an even better way than H2B-GFP.....?
Yes I do agree: GFP has been well-studied over the years. It has been shown that H2B-GFP system allows the high-resolution imaging of chromosomes, including the double minute chromosomes (DMs), without compromising nuclear and chromosomal structures.
As for the H2B histone protein, I would assume that it is located on the external side of the chromosomes (not contacting the DNA) and thus freely available and easier to tag. You can check that in the PDB perhaps. BTW, luciferase may work just as fine as GFP.
I guess that H2B is mainly used because it's the most represented canonical histone in nucleosomes during this step of cell cycle. In fact H2A is known to have a lot of variant, has H3. I don't know for H4.
Also I do agree with Vicente & Marceline, H2B-GFP is well-studied and used for many years.