Reliable automated counting of Nissl-stained cells in tissue sections may be difficult, since even the best ImageJ plugins will have difficulties in performing the segmentation of individual cells. This may be much easier with a staining method based on staining the nucleus. I personally would use DAPI/Hoechst to stain all nuclei (independent of cell type) and e.g. NeuN to detect neuronal cell nuclei (dual staining). Both signals are restricted to the nucleus, so the segmentation algorithms of ImageJ will have no problems and you can perform automated counting very easily (please countercheck the results by manual counting). If you are more interested in morphological features like axonal branching etc the situation becomes much more difficult.
Completely agree with Just Genius, pay also attention do do not have to thick (cell overlapping) and too low resolution pictures which will complicate the analysis.
Thanks to sanil sharma, just genius and Nicolas Casadei
Dear Just
Do you have any easy and applicable protocol for DAPI/hoecht staining? Also which brand of trinuclear not costly microscope and camera do you all suggest to take the images?
Both DAPI qnd hoechst come from lifetech. I attached the protocol suggested. We also use NeuN MAB377 from Millipore which is really good and could be bought already labeled with Alexa Fluor.
We use a Ziess axioplan with a CCD Ziess 503 mono for fluorescence. Maybe you can ask lab around yours if any microscope is available for this purpose.