A comparison of histochemical detection of GM1 ganglioside in cryostat sections using cholera toxin B-subunit after fixation with 4% formaldehyde and dry acetone gave tissue-dependent results. In the liver no pre-treatment showed detectable differences related to GM1 reaction products, while studies in the brain showed the superiority of acetone pre-extraction (followed by formaldehyde), which yielded sharper images compared with the diffuse, blurred staining pattern associated with formaldehyde. Therefore, the aim of our study was to define the optimal conditions for the GM1 detection using cholera toxin B-subunit. Ganglioside extractability with acetone, the ever neglected topic, was tested comparing anhydrous acetone with acetone containing admixture of water. TLC analysis of acetone extractable GM1 ganglioside from liver sections did not exceed 2% of the total GM1 ganglioside content using anhydrous acetone at -20 degrees C, and 4% at room temperature. The loss increased to 30.5% using 9:1 acetone/water. Similarly, photometric analysis of lipid sialic acid, extracted from dried liver homogenates with anhydrous acetone, showed the loss of gangliosides into acetone 3.0 +/- 0.3% only. The loss from dried brain homogenate was 9.5 +/- 1.1%. Thus, anhydrous conditions (dry tissue samples and anhydrous acetone) are crucial factors for optimal in situ ganglioside detection using acetone pre-treatment. This ensures effective physical fixation, especially in tissues rich in polar lipids (precipitation, prevention of in situ diffusion), and removal of cholesterol, which can act as a hydrophobic blocking barrier.