It was demonstrated that the combination of AFM and fluorescence can be a powerful tool for this type of project, where not only geometries but also identification of the imaged structures is critical. It was demonstrated that it is important to block non-specific binding of antibodies to the samples, and we are still fine tuning that process to optimize fluorescence imaging. A number of sample preparation methods have been tried to get as pure plasma membrane solutions as possible. We have successfully shown that the plasma membranes of the inner segments appear to have structured elements associated with them which have honeycomb-like structure. The identification of these structures with fluorescence proved more difficult so far, but ongoing work should optimize the appropriate conditions.
