We normally use copper grids covered with formwar for exosomes. Nickel grids are too sticky for common tweezers. Moreover, we use PSB/Glycine and PBS/BSA 1% for quencing (no TRIS, no acetylated BSA..too strong). No saponin is required. Protein A gold 10nm works very well to reveal HA tag. In the case of a rat Ab, I would recomment a bridge antibody (donkey anti-rat, rabbit anti-rat) and then PA gold.

Thank you Dr. Cortese! I am using nickel grids because I have read that they are preferred over copper for immuno incubations due to them being more inert. I may also end up wanting to do smaller gold particles followed by silver enhancement, which should be done on nickel grids I believe? I will definitely try the more mild buffers you suggested. As far as the saponin, we believe that our protein is inside the EVs, so Im using saponin to permeabilize the membranes. Is that not necessary?

For the antibodies, I realize I didn't update the protocols for the ones we are using. We are actually using a custom rabbit antibody for our protein of interests, followed by secondary nanogold (15 nm) conjugated goat anti-rabbit IgG. Thank you again for your response!

Hi Rachel,

Dr. Cortese's suggestions above are all very good. While copper grids will work for the experiments you describe, if you need to do silver intensification, nickel grids are probably your better bet, as nickel should not nucleate silver deposition.

For handling the grids, non-magnetic forceps may be very helpful. Nickel grids are magnetic, which can be a major pain when handling them with standard forceps. Non-magnetic forceps are available from any of the EM supply companies. If capillary action is a problem when moving grids from solution to solution, you might consider getting some anti-magnetic, anti-capillary forceps which have tips that meet at a wider angle than standard forceps (the two tips are not parallel to one another) and reduce the amount of liquid that moves up the forceps when picking up grids. These forceps aren't cheap, but they are worth it.

For the record, I have never labeled exosomes, but your immunolabeling approach seems fundamentally reasonable to me. Without knowing how much optimization you have already done, some things that you might consider adjusting would include (if you haven't tried all these things already):

1) Have you peformed a dilution series with your primary antibody to optimize it for this specific application? The optimal dilution for this application might be different from the optimum for LM immunolabeling or immunoblotting. Similarly, have you optimized the dilution for your secondary antibody? Back in the dim and distant past when I did some postembedding EM immunogold experiments on tissue sections, we often had to use gold-conjugated secondary antibodies at lower dilutions at the EM level than we would use for LM labeling with silver intensification.

2) Have tried longer incubation times in your primary antibody? Perhaps 1 hour at room temperature is not long enough to get optimal labeling. Similarly, have you optimized the incubation time for your secondary antibody (although I would guess an hour at room temperature should be sufficient)? Back in the dim and distant past when I had to do some postembedding EM immunogold experiments, we incubated our tissue sections in primary antibody for several hours at room temperature or for longer periods at 4 degrees C.

3) Have you tried using a smaller gold particle size on your secondary antibody? A smaller gold particle might improve penetration through the exosome membrane.

4) Have you tried slightly longer permeabilization of the exosomes? I imagine this is a tricky step to get the balance of permeabilization to expose the antigenic epitope without destroying the exosomes.

Good luck