This method combines the ability to stain for a membrane protein while also preserving membrane integrity and subcellular architecture. So it's really powerful for precisely localizing a membrane associated protein. The main advantage of this technique is that it combines the use of a robust genetically cloneable tag, APEX2, with state of the art methods in cellular preservation for electron microscopy, including cryofixation and freeze substitution.
Together, these allow precise protein localization within a well-preserved cell. This protocol can be used to dissect mechanisms by which viruses replicate inside the host cell. Viruses often will hijack host proteins during their infection scheme.
Viruses are the small genomes, which they utilize to replicate inside these host cells. Recently, we've been able to in unpublished data, label these proteins using this technology to understand how viruses replicate. This technique proved really critical in a recent collaborative study on a bacterial toxin that causes golgi breakdown.
Using cryoAPEX, we were able to precisely localize the toxin with the fragmented golgi pieces, something that was really difficult to see otherwise using normal immunofluorescence techniques. Demonstrating this technique will be Elaine Mihelc, a graduate student, and research associate, Stephanie Angel, from our laboratories. After seeding HEK293 cells in the dish, transfect the cells with APEX2 tagged mammalian expression plasmids using transfection reagent according to the manufacturer's directions.
At 12-15 hours post transfection, wash the cells once with PBS. Then, wash off the cells with PBS from the dish into a 15 milliliter conical tube. Centrifuge at 500 x g for five minutes.
Carefully remove the supernatant and re-suspend the pellet in two milliliters of 2%glutaraldehyde and 0.1 molar sodium cacodylate buffer at pH 7.4 at room temperature. Place the sample on ice to incubate for 30 minutes. Then, pellet the sample at 500 x g for five minutes at four degrees Celsius.
Wash the pellet three times for five minutes each with two milliliters of 0.1 molar sodium cacodylate buffer by centrifuging at four degrees Celsius. After centrifugation, remove the supernatant. To carry out the peroxidase reaction, wash the pellet by re-suspending in three milliliters of DAB solution followed by pelleting at 500 x g for five minutes.
Then, re-suspend the pellet in three milliliters of DAB solution with 5.88 millimolar hydrogen peroxide. Incubate for 30 minutes at room temperature. The pellet becomes visibly brown-colored, indicating the presence of the insoluble DAB reaction product.
After washing again with sodium cacodylate buffer in DMEM, according to the manuscript, re-suspend the cell pellet in 500 microliters of a cryoprotectant solution of DMEM containing 10%fetal bovine serum and 15%bovine serum albumin. Transfer the cell suspension to a 0.5 milliliter microcentrifuge tube. Pellet again, slightly increasing the centrifuge speed from 500 x g.
Discard the majority of the supernatant, ensuring that enough liquid is left so that the pellet will not dry out. Wick away any remaining liquid from the cell pellet using the corner of a laboratory wipe or paper towel. Let enough liquid remain that the pellet forms a paste, similar in consistency to toothpaste.
Aspirate two to three microliters of the cell pellet and deposit it onto a membrane carrier. Fill the well of the membrane carrier completely so that the surface tension creates a slight dome on top. Slide the membrane carrier into the cartridge and secure.
Place the cartridge into the HPF machine that has been prepped and primed and press start to freeze. Keeping the cartridges immersed in liquid nitrogen, remove the membrane carrier from its cartridge. Place into a plastic capsule and place the plastic capsule into a Cryovial full of liquid nitrogen.
In a chemical hood, prepare one milliliter per sample of the solution of 0.2%weight by volume tannic acid and 5%DI water in acetone in a Cryovial. Place them into liquid nitrogen to freeze. Place the freeze substitution mix one vials and the Cryovials containing the frozen cell pellets into the freeze substitution units sample chamber.
Transfer the inner capsule containing the membrane carrier from the liquid nitrogen vial into the corresponding vial containing freeze substitution mix one. Start a freeze substitution protocol at 90 degrees Celsius. After 24 hours, pause the freeze substitution and wash the samples three times for five minutes with acetone that has been cooled to 90 degrees Celsius.
Next, in a chemical hood, prepare one milliliter per sample of a solution of 1%osmium tetroxide, 0.2%uranyl acetate, and 5%DI water in acetone in a Cryovial and place them in liquid nitrogen to freeze. Place the Cryovials with the freeze substitution mix two into the freeze substitution unit and transfer the capsules from the third acetone wash into the freeze substitution mix two vials. Incubate them and freeze substitution mix two for 72 hours at 90 degrees Celsius, followed by gradual warming to zero degrees Celsius over 12-18 hours.
Keep the temperature at zero degrees Celsius and add pre-cooled acetone into the vials to wash three times for 10 minutes each. Prepare a resin mixture of choice in a plastic beaker according to the manufacturer's directions and add 2%4%and 8%of resin into the vials to immerse the capsules. Incubate for two hours each at zero degrees Celsius.
Then, incubate in 15%30%60%90%and 100%resin components of A, B, and D only for four hours each at room temperature. After the 20 hours, prepare a mixture of resin components A, B, C, and D and incubate for four hours. Place the membrane carriers with cell pellet side up into flat embedding molds and fill with resin mixture A, B, C, and D.Place them in an oven to polymerize at 60 degrees Celsius for 24-36 hours.
After polymerization, remove the blocks from the mold and place the sample in the vertical chuck of the ultra microtome where it can be visualized with magnification. Separate the membrane carrier from the block by a combination of dabbing liquid nitrogen on the membrane carrier to separate the metal from the plastic, and using a razor blade to chip away the resin around the membrane carrier. When separated, gently lift away the membrane carrier, leaving the cell pellet dome on the face of the block.
Place the block with the exposed cell pellet facing upward in a flat embedding mold that is slightly deeper than the first mold and fill with resin components A, B, C, and D.Polymerize at 60 degrees Celsius for 24-36 hours. Trim the block around the cell pellet using a razor blade. Then place the block in the sample chuck on the sectioning arm of an ultra microtome.
Using a glass or diamond knife, trim the block into a trapezoidal shape closely surrounding the cell pellet. Obtain 90 nanometer ultra thin sections of the cell pellet using a glass or diamond knife. Pick up a ribbon of sections on a TEM grid.
Dry the grid by blotting the edge on a piece of filter paper and store it in a TEM grid storage box. Mount the grid on the TEM holder and place the holder into the microscope. Use a Tecnai T12 at 80 kilovolts for screening cryoAPEX samples.
Acquire images of cells and subcellular structures of interest with APEX2 labeling. In this study, preparation of the sample by traditional APEX methods resulted in clear labeling of organized, smooth endoplasmic reticulum structures. At high magnification, the stacked membranes appeared ruffled and non-uniform gaps were present between concentric membrane densities, indicating poor membrane preservation and lipid extraction.
The sample prepared by cryoAPEX also had clearly-defined labeling of organized, smooth endoplasmic reticulum structures, however, the membranes were smooth and parallel and little to no lipid extraction was seen. TEM analysis of 90 nanometer thin sections revealed that Huntington yeast interacting protein E was present throughout the peripheral endoplasmic reticulum as well as the nuclear envelope. Additionally, the Huntington yeast interacting protein E density was resolved into regularly-spaced foci along the luminal endoplasmic reticulum membrane.
Huntington yeast interacting protein E distribution and foci were also visible in a sample prepared with traditional fixation and dehydration. However, extensive membrane disruption and extraction were present, making the sample suboptimal. APEX2 labeling was performed using three cellular markers, of which mito-V5-APEX2 provided specific staining of mitochondria only and CAAX-APEX2 produced distinct staining of the plasma membrane only.
No labeling was observed in intracellular organelles. Staining for the golgi lumen was also assessed using an Alpha MAN2-APEX2 marker as described in our original Sengupta et al manuscript. Following this procedure, three-dimensional methods such as electron tomography or focused ion beam scanning electron microscopy may be performed in order to investigate the 3D distribution of a protein of interest.
Many of the chemicals used in this protocol are toxic, so before use, safety data sheets should be consulted to ensure that the chemicals are handled using proper personal protective equipment and disposed of according to institutional guidelines.
