subcellular fractionation confirmed by western not consistent with microscopy - nuclear protein by microscopy not nuclear in fractionated sample -WHY? (Mar/15/2007 )

Hi guys,

I've fractionated some brain tissue using a kit from invitrogen which involves using different buffers and centrifugation speeds on the brain homogenate to fractionate the samples into a nuclear, cytosolic, cytoskeletal and membrane fraction. I confirmed fractionation had worked by running the fractions on a gel and western blotting for cytochrome C (membrane) NeuN (nuclear marker specific for neurons)and neurofilaments (cytoskeletal fraction). These gave the expected results - but when I blotted for my protein of interest - which by microscopy is both nuclear and cytoplasmic - it appeared in every fraction except the nuclear fraction.

my protein of interest is localised by its interaction with its target proteins - which are mostly nuclear

any ideas why I can't detect it in the nuclear fraction?

I'm writing my thesis at the moment - so any insight into this would be eternally appreciated !

Its quite difficult to have pure subcellular fractionation so it is always this semi pure fractionation.

What you visualise with microscopy is more reliable than western blot on subcellular fractionationated samples.

More over the way the fractionation is done can change things, like protiens moving out of the nucleus or displacement of dif. proteins.

Hi again,

Thanks for your thoughtful replies...

I have confirmed the nuclear location of my protein by both confocal microscopy (including z sections) and transmission electron microscopy. I agree that microscopy is more accurate for determining the localisation of the protein - but it would have been neat to get the fractionation to work too!

My protein of interest is known to move in and out of the nucleus depending on where the majority of its target proteins are residing, so I was thinking along the same lines as your comments about fractionation conditions somehow interfering with the location of my protein - for example the conditions could somehow prevent the interaction of my protein with its nuclear target proteins.

It would be nice to have some more specific ideas on why the fractionation process disrupts the nuclear location of my protein.

For example - my protein only interacts with its targets when they are phosphorylated so if the fractionation buffers somehow stripped the phosphate groups off the target proteins - this would explain why I cant detect my protein in the nuclear fraction (there is no evidence for this scenario and I did add phosphatase inhibitors to the homogenate before fractionation)

Are any of the conditions involved in frationation likely to disrupt protein- protein interactions?

Is there a specific chacteristic of protein that are resident in the nucleus that the buffer may be exploiting (like hydrophobicity of membrane proteins)? - if so as my protein that can shuttle in and out of the nucleus it probably doesn't share this characteristic

Sorry for the overlong post - not so much a post as a stream of consciousness!