Hardening mounting media that doesn't compress sample?

Are there hardening mounting media that really don’t compress the sample when hardening ?

Below are images of the same cells mounted with Vectashield, or Prolong Gold/Glass and only the cells in Vectashield look the same as alive. Slides were left to set overnight.

I don’t have pictures but Mowiol compresses the sample as well.

Any recommendations ?



Great pictures! Is the compression due to differences in osmolarity, that somehow causes the cells to deform? Is there something to do in the sample preparation that could mitigate some of it?

1 Like

Hi @oburri,
The cells imaged were only fixed with PFA before mounting the slides. However I noticed that the same happens when cells are permebilized with TX-100 so it shouldn’t be due to differences in osmolarity.

I would be interested in seeing someone else’s images comparing the same cells mounted with different mounting media.

Thanks !

1 Like

I’ve not come across a hardening media that doesn’t flatten specimens - it’s my understanding that components that cure/harden cause the flattening. I’m curious why you need the hardening? Long-term storage?

These images are great! Do you mind if we use them in our educational lectures? WIth full credit to you of course. If you don’t mind sharing them, please email me directly: jennifer@hms.harvard.edu.


I will echo Jennifer’s comments and her question - why do you need hardening? If you have a liquid mountant with a great refractive index that gives you awesome data - use it, if you are worried about your slips moving around you can always go old school and go around the edges with fast-set nail polish. If you don’t need anti-fade compounds in the moutant I’ve seen that some people get great results by mounting in immersion oil and then sealing the edges in nail polish, these obviously don’t keep long term, but the images are nice.

1 Like

Hi @jennifer and @JamesOrth,

I had 2 independent reasons why I was looking into hardening media:

  • I would like to align images taken on 2 different microscopes and I though adding TetraSpeck™ Microspheres to the sample to be able to align in XYZ including translation, rotation and scaling. And unless the beads are in hardset mounting medium they move around.
  • I wouldn’t need any more to deal with nail varnish; which is time consuming when you have a lot of coverslips.

Happy to send you the images by mail.

Thanks !

Those are great reasons. For doing something related to what you describe I’ve had good success using gridded coverlips or scratching my own coverslip with a scribe into a pattern that is recognizable, here I take a low mag image to see where all of the scratches are so I can take to another microscope, orient, find my area, then bring up the mag. This is great for XY, not in Z. I sort of miss the days of 100s of slips. Best wishes with your experiments. If I come across anything about hardening-non compressing mounting agents I’ll be sure to post it.

1 Like

Hi LPUoO - After a quick search I see there is a mounting agent called FluorSave (MilliporeSigma) that purportedly hardens in about an hour, and “shrinkage after one week can cause tissue damage”, I wasn’t able to find the original reference that shows this. It is inexpensive, could be worth a try, mount, wait an hour, image. Good luck.

Regarding the microspheres, an alternative to a hardeset mounting media would be crosslinking the beads to the slide/cells. But usually unless they are specially designed for that it is impossible to crosslink that sort of fluorophores.

I have a nice slide that I got from GE when they were giving an OMX demo about hardening vs non-hardening (#8 in the following presentation):
Feel free to use it as long as you attribute the person from GE that gave it to me.

The only thing I don’t like about that slide is that it doesn’t have a side view of what Prolong-Gold/Diamond looks like when you seal a coverslip with nail polish immediately after mounting (ignoring what the manual says). This keeps the Prolong from drying and squishing the sample, as well as the RI from going up. Despite the mismatch in RI and the concomitant spherical aberrations deeper in, we find that it is better to not flatten the sample for anyone interested in imaging in 3D. If you don’t care about 3D structure, flattening the sample is great for widefield. So, what we recommend if using a hardening media like Prolong is seal immediately if you are doing confocal, let cure if you are doing widefield.

What my users would really like is a commercial product that does not harden, has good antifade properties and a high RI (matched to the 1.518 of typical Big 4 oils). Or, something homebrew that is cheap and easy to make, with the same characteristics. Some clearing solutions like TDE or SeeDB2 variants would fit the bill, but I don’t have a good idea of the cost or hassle level involved in making them.

1 Like


great test and beautiful images!
I performed a similar test some while ago and found ProLong Diamond (hardening) performing the best in this test. (see pictures). Very recently, I also tested ProLong glass and shrinkage was also apparent there. However, the deformation that you demonstrate was not visible for me, but it is a valuable information.


Dear Pablo,
Thank you for your ppt. I will use your picture in our educational ppt in our platform in Belgium. With full credit to you of course.

Dear bzobiak,
Can I use your pictures? For our educational ppt in our imaging platform. With full credit to you of course.

Note that Dapi staining is not necessarily sufficient to detect in-foldings created by sample compression. In my opinion nuclear envelope staining (e.g. Lamins) are the best to see those compression artefacts.


I thought of this thread again after watching: https://www.youtube.com/watch?v=t-xBQ8buZlE&list=PLN-QyZNMh3PvM0FQg_3Cjnc8XLJHs4Efn&index=3
I’m a bit curious how using fixed samples with a deep learning algorithm might impact the estimated morphology in live cells. Especially since I probably wouldn’t have considered the mounting media if I had started doing something similar; what else might play a factor?

At a more pertinent time point.

Have you tried SlowFade Glass?