Dear all,
I am interested in an mRNA that has been described as being bound to the
cytoskeleton. There are two published papers showing that a labelled
fragment of the 3' UTR can be crosslinked to various (unidentified)
proteins. I am hoping to identify the proteins and characterise them
further.
Unfortunately I am having great difficulty in replicating the original
findings. In our hands the UVXL experiment only gives very low intensity
signal, the band sizes are different to the published work (though I note
that the two published papers also reported a range of band sizes!), and
in our hands the binding cannot be competed by antisense oligos, so it's
not a sequence-specific interaction).
The UVXL binding assay uses a protein extract made with a cytoplasmic
extraction buffer (CEB). Then, a labelled RNA probe (e.g. DIG or biotin
labelled) is added, allowed to interact with the proteins in the presence
of competitors (e.g. heparin, tRNA, other irrelevant unlabelled RNA). This
binding step also uses CEB as the buffer. After binding, the RNA is
UV-crosslinked to the protein, followed by RNase treatment to remove
unbound probe. Finally, Laemmli buffer is added and the whole lot goes on
a Western blot.
CEB composition: 10 mM Hepes, pH 7.5, 3 mM MgCl2, 14 mM KCl, 5% glycerol,
0.2% Nonidet P-40, 1 mM dithiothreitol with freshly added protease inhibitors.
Question 1: Surely this buffer will only extract cytosolic proteins, and
not the cytoskeleton? If the mRNA really is cytoskeletally-bound, then
potentially the binding seen in the original papers represents binding to
cytoskeletal contaminants leaking into the cytosolic fraction?
Question 2: Will CEB lyse nuclei and/or other organelles, and does this
depend on the method of homogenisation? We have been using a rotor/stator
homogeniser, but the original papers don't say what was used. How will the
homogeniser compare to grinding frozen tissue in a mortar and pestle, say?
Question 3: If indeed the binding proteins of interest are poorly
solubilised (or not solubilised at all) in CEB, what are the options? If
I try more aggressive buffers such as RIPA, or varying levels of Triton
in CSK, will that compromise the subsequent binding reactions? Should I
perhaps make an extract in RIPA, then dialyse back to CEB before doing the
binding reaction?
Question 4: Could I even simply lyse the cells in CEB _without_ spinning
down the insoluble material, and use this for the RNA binding reaction,
complete with the suspended insoluble/cytoskeletal/membrane fragments? I
know UV crosslinking is sometimes performed on intact cells to crosslink
endogenous mRNAs to proteins. As long as the outer cell membrane gets
disrupted to allow the labelled RNA probe access to the cell contents,
then it may be able to bind to its protein partners.
Many thanks,
Peter Ellis
_______________________________________________
Methods mailing list
[email protected]
http://www.bio.net/biomail/listinfo/methods
