I'd like to compare the nuclear from the total RNA fraction, especially for small RNA. Is there any good protocol? How to be sure about the integrity and quality of the RNA?
When comparing cytoplasmic (or total RNA) with nuclear RNA, it is crucial that you don't standardize the RT reactions using the same mass input amount of RNA. Instead, you should use the same volume of RNA (extracted from the same number of cells) as input in the RT. As RNA extraction is a somewhat variable process, we recommend replicated RNA extractions from parallel cell cultures to assess reproducibility of the results. You also may want to include control genes of which you know that these are predominantly expressed in the nucleus vs. cytoplasm (and vice versa).
Because nucleus and cytoplasm contain different amounts of RNA. If you use the same amount of RNA in your workflow, you miss this difference! I now also recommend to add spike-in RNA to the lysates to control for RNA purification efficiency and normalization.