What is a linker histone doing there?

By way of introducing this short entry:  as is probably true for most blogs that discuss various and sundry aspects of science, I have tended to focus on reviews or peer-reviewed research papers – “the literature”.  There is, however, a whole lot more to the lab than these finished and polished products.  What I want to do with this entry is a bit different.  Instead of talking about a complete study, I thought I would talk (briefly) about some results from my lab that, for various reasons, never found their way into print.  Ideally, someone will read one of these essays and speak up, telling me just what is going on and how it fits in with other data or models.

The following is one such example, a result that is curious and perplexing.  I chose it because it comes with pretty pictures, and because it is a segue for another essay that I will post in the future.  The data is from a thesis of a student of mine – Kevin Forbes.  The experiment itself is 7-10 years old (I have forgotten just when this study was done), and I made sure that Kevin would be OK with this before I posted anything.

For reasons that are apparent in his thesis (Fig. 2.8), Kevin got interested in a novel protein that bore an unmistakable resemblance to a linker histone (an H1-like protein).  Since he was studying factors that affect the activities of poly(A) polymerases, he reasoned that this protein should co-localize in the nucleus of plant cells.  So he did the usual experiment – fuse the coding sequence of the H1-like protein to RFP, transiently express the construct in tobacco cells, and record the subcellular distribution of the fluorescent protein.  This is what he found:

He co-expressed a poly(A) polymerase -GFP construct in the same cells:

As expected, this protein decorated the nucleus of the transfected cell.

The merged image:

(The small red structures are chloroplasts, that are apparent in the merge due to autofluorescence.)

For everyone’s information, here is how GFP (A) and RFP (B) by themselves distribute in tobacco cells:

What we did not expect at the time, and what we never followed up on or explained (Kevin had other projects that had better prospects for publication) was the observation that the H1-RFP fusion protein decorated the plasma membrane so clearly and specifically.

Any ideas?  Readers are invited (begged) to chime in with thoughts, perhaps even answers.