Simple, Scalable Proteomic Imaging for High-Dimensional Profiling of Intact Systems [full article]

Evan Murray, Jae Hun Cho, Daniel Goodwin, Taeyun Ku, Justin Swaney, Sung-Yon Kim, Heejin Choi, Jeong-Yoon Park, Austin Hubbert, Meg McCue, Young-Gyun Park, Sara Vassallo, Naveed Bakh, Matthew Frosch,, Van J. Wedeen,  H. Sebastian Seung, and Kwanghun Chung. Simple, scalable proteomic imaging for high-dimensional profiling of intact systems, Cell, Dec 3:163(6): 1500-14. doi: 10.1016/j.cell.2015.11.025. PubMed PMID: 26638076.

The SWITCH tissue processing framework consists of three key methods:

• Scalable and uniform tissue preservation
• Easy and rapid tissue clearing
• Virtually unlimited rounds of antibody labeling, stripping, and relabeling.

With these methods, you can perform many labeling rounds on the same tissue. Better yet, this is scalable to large volumes.

The central idea is SWITCH -- a reaction control strategy that allows for uniform tissue processing. Basically, you allow chemicals to diffuse into the tissue first without reacting with the tissue (this is called SWITCH-Off). Then, when they are fully dispersed, you let them react (this is called SWITCH-On).

For example, imagine you were doing whole brain antibody labeling. You incubate the brain in antibody solution for a week, hoping that it's enough. But when you image, you will find that only the surface and regions close to the surface are stained, while the core is not stained very well.

The illustration above compares conventional labeling to SWITCH labeling. As you can see, if you do the SWITCH approach and give enough time for antibodies to diffuse throughout the tissue, you will get a uniform staining throughout the tissue.

This concept applies to fixation as well. It turns out that fixation is very similar. The surface always reacts first, then the core. But SWITCH protocol allows you to do uniform fixation. This becomes very useful when, for example, you are working with tissues that you can't perfuse (perfusion bypasses the diffusion step because it disperses chemicals through vasculature) or tissues from tissue banks.

Fixation

We needed a chemically and mechanically stable tissue to enable multiple rounds of labeling. At the same time, we wanted to be able to clear and label the tissue afterwards for intact tissue imaging. So after testing out different fixatives, we developed an optimized fixation solution (4% paraformaldehyde [PFA], 1% glutaraldehyde [GA]) that best preserved antigenicity and provided chemical and mechanical stability.

Fixation Protocol

Perfusion fixation

If it is possible, perfusion is the preferred method of tissue preservation. These are the materials that you will need for perfusion fixation:

Create a solution with a final concentration of 1X PBS, 4% paraformaldehyde (PFA), and 1% glutaraldehyde (GA). As 40 mL of this solution is necessary for each perfusion, a typical recipe is: 4mL 10X PBS, 5 mL 32% PFA, 0.8 mL 50% GA, and 30.2 mL water. This solution should be made fresh immediately prior to performing perfusion and kept on ice at all times. It is recommended to chill all of the separate ingredients before mixing the components.

Using the perfusion technique of your choice, first perfuse 20 mL of ice-cold PBS through the beating heart of an anesthetized mouse, followed by 20 mL of the ice-cold perfusion solution described above. Take care not to introduce any bubbles during the procedure, and use a flow rate slow enough to avoid damage to the vasculature or brain sample (<5 mL/min). After both solutions have been perfused, carefully remove the brain from the skull using any technique you are comfortable with. The dura membrane should also be removed during the process. Place the sample into 20 mL of perfusion solution and incubate at 4 ˚C with gentle shaking for 3 days.

SWITCH fixation

You should use SWITCH fixation if you can't do perfusion with the 4% PFA, 1% GA solution.

SWITCH fixation involves two steps: SWITCH-Off at pH 3 to disperse the fixatives (fixatives don't work well at pH 3) and SWITCH-On at pH 7 to begin fixing the tissue.

These are the materials that you will need for SWITCH fixation:

If you are using PFA-fixed tissues (e.g. tissues from tissue banks), you need these two solutions:

• SWITCH-Off Fixative: 1X PBS, 4% GA, 0.05 M KHP, titrated with HCl to pH 3.
• SWITCH-On Fixative: 1X PBS, 1% GA.

Tips for making the SWITCH-Off Fixative: Titrate a bottle of PBS to pH 3 using HCl. Create solutions of 0.1 M HCl in water and 0.1 M potassium hydrogen phthalate (KHP) in water. Finally, mix these solutions in a ratio of 2:1:1 (pH 3 PBS):(0.1 M HCl):(0.1 M KHP). To this new solution, add a stock solution of GA to make a final concentration of 4% GA. Ensure that this solution stays cold at all times. It is recommended to chill the solution before adding GA.

You need to add the KHP to the SWITCH-Off Fixative to undo some of the PFA fixation.

(If you are using fresh tissues (not fixed previously), we recommend you first fix with PFA for several days before proceeding. This is because tissues can degrade quickly and because PFA fixation is quick and does not require SWITCH processing.)

Once you're ready, incubate the sample in 40 mL fixation-OFF solution at 4 ˚C with gentle shaking for 2 days. The sample should then be moved to fixation-ON solution at 4 ˚C with gentle shaking for an additional 2 days. (The timing for the Fixation OFF and ON steps is dependent on the sample size and may need to be optimized from these starting values on a case-by-case basis. We found that these parameters worked well for banked human samples of roughly 0.5-1.0 cm thickness.)

Inactivation and Clearing

Inactivation Protocol

Once you have a SWITCH tissue (4% PFA and 1% GA fixed tissue), you need to inactivate the tissue. Inactivation means terminating any fixation reactions still going on within the tissue. If you don't do inactivation, you might find a lot of background signals after labeling because antibodies become fixed to the tissue.

For inactivation, you will need:

Make a 4%(w/v) acetamide, 4%(w/v) glycine, 1X PBS solution and titrate it to pH 9 with NaOH.

Now you're ready to inactivate some tissues. But before you do that, you want to make sure the tissues are free of excess fixatives. You don't want to start fixing glycine and acetamide everywhere.

So, after fixation via either perfusion or SWITCH, the sample must be washed in PBST to remove unbound fixative molecules. For mouse brains, 2 washes of 6 hrs each at RT with gentle shaking was sufficient.

Then, finally, to inactivate remaining fixative molecules, you can incubate the sample in inactivation solution at 37 ˚C overnight (O/N for short). Then check the sample in the morning. If the solution turned yellow, the inactivation solution should be replaced with fresh solution and the sample incubated for several more hours. (If the sample needs to be cut, this should take place now before the sample is cleared.)

Clearing Protocol

We clear tissues with SDS. Briefly, SDS goes into the tissue and takes out the light-scattering lipid bilayers, making the tissue more transparent. This can be a time consuming step for large tissues.

One way to speed it up is to use a warm SDS solution. We speed it up more by using a hot SDS solution. Thanks to the stable SWITCH fixation (4% PFA, 1% GA), the tissues remain pristine even at high temperatures.

One problem though -- the tissue becomes brown at high temperatures. This is because of the Maillard's reaction, the same reaction that turns buttered bread into a golden-brown toast. We overcome it by using sodium sulfite, an antioxidant known to suppress this reaction.

For clearing, you'll need:

The clearing solution consists of 200 mM SDS, 20 mM sodium sulfite, 20 mM boric acid, and 10 mM sodium hydroxide. The pH should be around 9. Read this for more details.

Inactivated samples must next be incubated in clearing solution to wash away remaining inactivation solution and to distribute sodium sulfite through the sample. After 2 washes of 6 hrs each, the samples should be placed in a tube of fresh clearing solution, which should then be placed in a water bath heated to 70 ˚C. Other temperatures or methods of consistent heating may be used, but samples may deteriorate over time at higher temperatures. If a sample contains fluorophores that were genetically-encoded, introduced through viral injection, etc., then the sample may be cleared at 37 ˚C to preserve this fluorescence. The clearing process will take much longer at this low temperature, but temperatures higher than this will result in loss of fluorescence during clearing. Falcon tubes can become fragile over time in these conditions, so it is necessary to frequently check that the tubes have not begun to leak.

Labeling

We extensively use the SWITCH strategy for large volume labeling. Make sure to read, read, and read the section above before you start doing any large volume labeling! Here we will provide two protocols: one for myelinated fiber labeling and one for antibody labeling. You can use these as a starting point for your labeling adventures.

Myelinated fiber labeling protocol

You will need the following reagents:

Make the following solutions EXACTLY:

• SWITCH-Off: 10 mM SDS in 1X PBS (dissolve some SDS into 1X PBS)
• SWITCH-On: 1X PBS, 0.2% Triton X-100 (aka PBST; dissolve some Triton X into 1X PBS. We recommend you make a 10% Triton X stock to let it dissolve faster)

We take advantage of the fact that DiD staining doesn't work at 10 mM SDS. So we let the DiD diffuse into the tissue with SDS (SWITCH-Off), then switch the buffer to one without SDS, to label the myelinated fibers (SWITCH-On).

Myelinated fibers can be readily visualized with the lipophilic DiD fluorescent molecule. The sample should be equilibrated in a solution of 10 mM SDS in PBS in order to distribute SDS molecules throughout the sample. The sample should then be placed in a volume of DiD-OFF solution just large enough to cover the sample and incubated at 37 ˚C with gentle shaking for 12 hrs to 7 days depending on the size of the sample (1 mm-thick section to whole mouse brain, respectively). The sample should then be moved to 40 mL of PBST and incubated at 37 ˚C for 12 hrs to 2 days (same deal). We have also observed that tomato lectin and nuclear stains such as DAPI or Syto16 can be used with this SWITCH approach.

Antibody labeling protocol

You will need the following:

You will also need some antibodies.

Again, make the following solutions:

• SWITCH-Off: 0.5 mM SDS in 1X PBS
• SWITCH-On: 1X PBS, 0.2% Triton X-100

You can add more SDS if you find that 0.5 mM SDS is not inhibiting your antibody well enough. You can add up to about 10 mM SDS, and then there's no additional benefits. This is because only the monomeric form of SDS inhibits binding, and the maximum amount of monomeric SDS is about 10 mM -- beyond that point (called the critical micelle concentration), the SDS will exist as a mixture of micelles and monomers (e.g. 200 mM SDS has 190 mM micelles and 10 mM monomers).

Onwards: the sample should be equilibrated in antibody-OFF solution in order to distribute SDS molecules throughout the sample. The sample should then be placed in a fresh volume of antibody-OFF solution just large enough to cover the sample, and then antibodies should be added in the desired proportions. The sample should then be incubated at 37 ˚C with gentle shaking for 12 hrs to 7 days depending on the size of the sample (1 mm-thick section to whole mouse brain). The sample should then be moved to 40 mL of PBST and incubated at 37 ˚C for 12 hrs to 2 days.

Delabeling

You can strip the antibodies for multiple round labeling by incubating the sample in the clearing solution at 70 ˚C for 2 hours to O/N depending on the size of the sample. Basically, think of it like you're clearing again, except you're clearing away the antibodies. You can then label again, following the labeling protocol. It's like a coloring book -- color one color, take a picture, erase, and color again.

Optical clearing

The tissue is transparent, but not transparent enough for imaging. Before you can image your labeled samples, you should optically clear the tissue by incubating it in a refractive index matching solution. We'll write another post that explains how this all works, but for now, you can follow the following protocol for our custom refractive index matching solution.

Our optical clearing solution requires:

Make a 23.5% (w/v) n-methyl-d-glucamine, 29.4% (w/v) diatrizoic acid, and 32.4% (w/v) iodixanol solution. Basically, add 40 g n-methyl-d-glucamine, 50 g diatrizoic acid, and 55 g iodixanol to 100 mL of water. Or, if you're using the 60% iodixanol solution from Sigma, add 40 g n-methyl-d-glucamine, 50 g diatrizoic acid, and 92 mL of 60% iodixanol to 63 mL of water.

Do not use heat when mixing the solution, as this will cause a color change. This solution should be stored carefully to ensure that no water is lost, as just a small amount of evaporation will result in precipitation. Teflon tape can be used to increase the security of the bottle’s seal, and parafilm can be used around the cap.

The sample should be washed in optical clearing solution at least 3x for 6 hrs each at 37 ˚C with gentle shaking. After the final wash, the sample should be clear enough to easily see through by eye. If the solution immediately surrounding the sample seems inhomogeneous, it suggests that the sample has not yet fully equilibrated with the solution.

SWITCH Video Gallery

Protocol

The full protocol can be downloaded here.

Discussion

We want this to be a two-way discussion. Please post any questions or comments below.