Glutaraldehyde/OsO4 protocol
  1. Remove monolayer culture from incubator and rinse briefly in protein-free Ringer’s solution
  2. Primary fix for 30-60 min at RT* in 200mM (2%) glutaraldehyde in “NaHCa” buffer*
  3. Rinse 2x in NaHCa buffer (5min each)
  4. Mordant with 1% tannic acid (plus 0.075% saponin, to help the tannic acid get into cells) also in NaHCa buffer, for 15 min @ RT
  5. Rinse 2x in NaHCa buffer (5min each)
  6. Rinse 2x in 0.1M cacodylate buffer pH7.4 (10 min each)
  7. Postfix 20 mM (0.5%) OsO4 in 0.1M cacodylate buffer pH7.4, for 15 min @ RT
  8. Rinse 2x in 0.1M cacodylate buffer pH7.4 (5 min each)
  9. Rinse 2x in 0.1M Na acetate buffer pH5.2 (10 min each)
  10. Block stain with 4% uranyl acetate dissolved in 50mM Na acetate buffer pH5.2, for 15 min @ RT (covered with Al foil to keep dark)
  11. Rinse 2x in 0.1M Na Acetate buffer pH5.2 (5 min each)
  12. Dehydrate through 25%, 50%, 75%, 95%, and 100% ethanol (20 min total for this)**

*Aldehyde fixaton should always be done at room temperature (RT), never in the cold or in the ‘frig, since it’s not fixing the lipids in biological membranes, and chilling lipids promotes artifactual phase-separations.

*"NaHCa" buffer consists of 30 mM HEPES buffer, brought to pH 7.4 with NaOH (not KOH!), with 100 mM NaCl present to make it roughly isotonic and 2 mM CaCl2 present to preserve cell membranes.
(P.S. For virus experiments, 10mM MgCl2 is also added to this fixative, to improve the preservation of delicate & stressed cytoplasmic proteins & cytoplasmic polynucleotides.)

(In making up the buffer solution, to which the glutaraldehyde will be added later, be sure not to add the CaCl2 until after pH'ing the buffer to 7.4, or else the calcium will precipitate when the NaOH hits the surface of the solution.)

(We usually add the CaCl2 from an unbuffered 1M stock in distilled water.)

(When adding the fixative itself to this buffer, use an "EM grade" of glutaraldehyde, usually supplied by commercial companies as a concentrated solution (25-70%) in a sealed 2ml or 10ml vial over argon or another inert gas.)

**(Actually, samples probably should never see lower than 50%ethanol, and should never be rinsed in distilled water, because in both cases they tend to swell.  It’s bad enough that they are going to shrink in the higher alcohols.) 

  1. Embed in epoxy resin according to the following patented “protocol for fast embedding”, which involves diping the coverslips sequentially into nine small beakers containing the following:
  1. 100% Ethanol (really dry!)
  2. 100% Ethanol, again
  3. 100% Ethanol, again
  4. 100% propylene oxide
  5. 100% propylene oxide, again
  6. 50/50 (V/V) propylene oxide and epoxy resin*
  7. 50/50 (V/V) propylene oxide and epoxy resin, again
  8. 100% epoxy resin
  9. 100% epoxy resin, again (The coverslips should of course be held with forceps
    and should be ‘swished’ around in each beaker for just about 30 seconds)

(*The full mixture of epoxy resin, including the accelerator, is made up fresh right before use, and is outgassed by repeated vacuum-application many times, until all the bubbles come out of it.  This we do with the vacuum-oven that will be used for the final epoxy polymerization, but with the heater turned off, so the epoxy doesn’t get heated and start to polymerize.  This the most tedious step of the whole procedure..You’ll see!))

The coverslips are then set in aluminum dishes filled with a ~1mm deep layer of epoxy resin and placed in the vacuum oven (now warmed up to 65degC), and evacuated two or three times, to draw off any residual propylene oxide.

Finally, the coverslips are inverted over  “Beem” capsules (or large gelatin pills), held perfectly upright somehow, and filled to the very top with the same fresh epoxy resin (trap no bubbles!), and placed in the heated 65degC vacuum-oven for polymerization. (Don’t apply any vaccum at this point, though, or bubbles might get trapped under the coverslip.)

After polymerization, the coverslips are “popped” off the Beem capsules by the usual technique: touching the glass to a copper or brass block partially immersed in LN2, and then pulling it off the epoxy.

(Actually, we abhor this technique for separating coverslips from polymerized epoxy, and have switched completely to growing all our monolayer cultures on Thermanox® coverslips (15mm diameter circles).  These have adhesive properties almost identical to Falcon tissue culture dishes (much better than glass), and much more importantly, they peel off the “Beem” capsules without the slightest problem, and without all the LN2 long as one removes the sample from the 65degC oven and peels the plastic coverslip off the capsule immediately, e.g., while it is still hot!!

If you don’t know about these Thermanox coverslips, you can read up on them at the following links:
The only problem with Thermanox coverslips is that even though they’re clear and transparent, they have very poor optical properties, so you cannot image cells on them by looking through them,  but must use a water-immersion objective in the LM, so as to image cells on them from above.