Well, to start with, doing a fractionation on a frozen tissue is not the best idea, since the freezing process itself is disrupting cellular morphology and cellular structures due to the volume expansion of crystal water.

Anyway, classical cytoplasmic/nuclear fractionation employs a first gentle lysis buffer (typically 1% tritonX-100 or similar detergents and 100-150 mM NaCl in some buffering medium - tris/HCl or phosphate or sulphonates such as hepes - EDTA and protease inhibitors) in a glass/teflon douncer, pelletting nuclei at 10000 rpm, washing the pellet in the lysis buffer at least once, and finally lysating the pellet in Laemmli buffer. Keeping in mind that nuclear proteins are in general around 10% of total cellular protein content. And working all the time on ice, as always with proteins.

Good luck!

I would recomend the use of g instead of rpm as at 10000 rpm depending on the rotor diameter can pellet the synaptosomic/mitocondrial/microsomal fraction that could alter the results of your study. After the glass/teflon douncer step with a regular lysis buffer as the one previously suggested, 1000 g 10 minutes will result in cell debris and nuclear pelleting. 1000 g centrifugation supernatant could be considered as cytoplasmic proteins (or at least non nuclear proteins) but keep in mind that in that supernatant you will have the synaptosomic/mitochondrial/microsomal fraction.

I hope it will be useful.

Good luck.

Thanks to you both for your comments! Unfortunately given that the samples will be collected after behavioral training there's no way to use fresh tissues for this. So thats why I'm really hoping to get this working on frozen samples. I will keep it in mind that fractionation from frozen samples isn't common so using fresh will require me to rethink about my experiment design.

Andres has a point, 10000 rpm could be too high for brain tissue. Although synaptosomes and mitochondria are lysed in 1% TritonX, (and microsomes in Triton should not really form and anyway at 10000 rpm will remain in solution, in standard fractionation preparations without detergent you need 100000 g to pellet them), the PSD and pre-syaptic matrix are insoluble in TritonX and will at least partially precipitate at 10000 rpm contaminating your nuclear fraction. So probably the best would be spinning between 1000 and 3000 g.

If you can lysate quickly your brain tissue as soon as you sacrify and dissect the animals after behaviour, I believe it will be more reliable. Maybe you can organize with a colleague to work in tandem??

Thanks Pietro. I'll think about trying to quickly lyse the tissue following behavior. Thanks again! This is great information!

I highly recommend a product from Pierce (Thermo Scientific) called NE-PER

http://www.piercenet.com/product/ne-per-nuclear-protein-extraction-kit

I was able to reduce the amount of tissue I needed significantly when using this kit over our lab protocol. In addition it was done in a couple hours rather than a whole afternoon. I would at least see if they will give you a free sample to try out.

Thanks Michael! Have you used this kit on frozen brain punches before?

I did not use this kit on frozen tissue but instead for primary cultures of human macrophages harvested from lung lavages. As Pietro stated you will have some nuclear protein contaminating the cytoplasmic fraction no matter what protocol you use coming from frozen tissue. Brain tissue homogenizes very easily so it is just the ice crystals poking holes in the nuclear membrane that you have to take into consideration when interpreting the cytoplasmic protein levels.

Hi Stephanie, thinking it over a bit, and stimulated by the discussion and proposals, I might have an idea: if you are freezing punches, so quite small tissue pieces, you could probably succeed by vitrifying them, just dipping them into liquid nitrogen, and then storing them also in liquid nitrogen or in its vapour. The freezing should be fast enough to avoid water crystallization if the punch is tiny, 1-2 cubic mm.

In any case I would reccommend testing your fractions with markers of the compartments you are separating, so, for example, if your aim is protein analysis I would run a WB for some histone and maybe also RNA Pol II, for example (and a cytoplasmic marker as well, for reverse contamination). In the end perfect separation is probably impossible, but if your enrichment is >90-95% it will probably be perfectly acceptable for most applications.

can you please tell me how to isolate rough ER fraction from rat brain.

i have found a very nice protocol for cells that works like magic with no cross contamination and I'm going to try it on frozen tissue as a test knowing that it would not be efficient like what i get on cells but worth trying I will update here :) https://rockland-inc.com/NuclearExtract.aspx

Sepideh Parsi Is this kit good for human brain fractionation too???

I am trying to do fractionation of frozen human brain tissue for nuclear and cytoplasmic fractions, but the yield is too low. I used Biovision Kit, with mouse tissue, the yield was good though.