LCM histology Problem - Histology is no longer discernable after fixation. (May/27/2008 )

I work with mouse lung and have found an issue in the LCM protocol used in other tissues. Every time I dehydrate the lung after staining, the individual cells shrivel up and the morphology becomes so poor that I can no longer identify cell types. I currently flush the lung with PBS, followed by 50% OCT with 10% Sucrose prior to flash freezing. From there, the lungs are placed in a microtome, embedded in OCT and sectioned. The sections are stained with H&E and then dehydrated with ethanol. I can watch the tissue lose morphology as the ethanol dries. The addition of sucrose greatly helped, but it is not sufficient. Would a non-crosslinking fixative help in this regard? Are there any other tricks to improve the morphology further.

It will depend on what you want to do with your sections... Are they for immunocytochemistry (if so, please specify if you are going to use peroxidase or fluorescence protocols) or just for histology?