Exercise in tissue engineering, Halloween decoration, or mad scientist show piece? Take your pick...

Some people like to make their Halloween decorations from papier mache. We used raw, bleeding pig's hearts!

Last year, we tried to decellularize some chicken hearts and gizzards at BioCurious for Halloween. This year, we tackled a full-size pig heart, with members of Counter Culture Labs, our new DIYbio group in the SF East Bay, with some help from friends at BioCurious.

Decellularization is a tissue engineering technique designed to strip out all the cells from a donor organ, leaving nothing but the connective tissue that used to hold the cells in place. This scaffold of connective tissue - called a "ghost organ" for its pale and almost translucent appearance - can then be reseeded with a patient's own cells, with the goal of regenerating an organ that can be transplanted into the patient without fear of tissue rejection. A decellularized mouse heart regenerated with human heart precursor cells will actually start beating autonomously! Here's some amazing videos on this work by Doris Taylor at U. Minnesota:

Creating a beating heart in the lab
Surprising Beauty, Holding A Pig's Heart

Getting all the cells to coordinate correctly in a large heart is still a big challenge. That part is still years away from clinical use. We're just doing the first part though: decellularization of a pig heart. Why? Well, what self-respecting mad scientist wouldn't want to have a Ghost Heart in a Jar!

If any of this seems way too challenging to you - note that none of us had ever done anything like this before! We just read up on what it took to do this, figured out where to get a pig heart, got a few pieces of equipment, ordered some chemicals, and got to it! The results look absolutely spectacular, and we learned tons along the way. Yes, this project is a bit complex, and there's lots of long words in the steps below - but don't let that scare you away! The whole thing can be accomplished in a single day, and we'll tell you about DIY alternatives for all the equipment and chemicals used here.

The decellularization is achieved using enzymes and detergents to break open the cells and flush out the cellular content. The solutions need to be perfused throughout the heart muscle using a pump. Not just through the chambers of the heart, mind you, but actually through the coronary arteries that provide blood supply to the muscle of the heart itself. The published protocol takes approximately 10 hours, but since we are planning this as a show piece and not to reseed them with cells, we have taken some shortcuts.

Please use appropriate safety equipment and measures. One of the nice aspects of this experiment is that all the chemicals you're working with are fairly non-toxic. But you're also working with lots of fluids under pressure, and you really, really don't want assorted pig heart juices squirting into your eyes - trust me on that! So make sure to wear some goggles at least. A lab coat to protect your clothes from giant blood stains wouldn't hurt either (unless you're into that kinda thing).

If you use something like an aquarium pump please note the problems the solutions used could cause the pump, as well as contamination issues. A fountain pump from the hardware store might be another interesting alternative for high volume and relative high pressure. But most fountain pumps are immersion pumps, and are *not* designed to handle salty, conductive water - try at your own risk!

Check out the timelapse video we made we made of the whole procedure:
 

If you hunt around, you can probably find some pig hearts for sale in a Chinese supermarket, or from an adventurous butcher. However, hearts that have been butchered for food typically have the arteries at the top of the heart cut off, and a slash through the chambers of the heart to let the blood drain out. Unfortunately, that makes them entirely unsuited for what we need.

What you need is a pig heart that has been butchered specifically for biological experiments - what's called a "bio-heart" in the trade. The good folks at the California Academy of Sciences recommended C&M Meat Company, a meat wholesaler in Berkeley that should be able to get us a bio-heart fairly cheap.

Unfortunately, when you're trying to get a single heart from a wholesaler... you're not exactly a high priority for them. They agreed to get us some bio-hearts, but only if we ordered at least five at a time. So after getting the runaround and wasting a couple more weeks of our time, we finally received a surprisingly neat cardboard box, containing five very raw, and very bloody pig hearts!

Our bio-hearts came with about a foot or more of extra arteries attached. You'll need no more than about an inch of the aorta to attach a hose for the perfusion, so you may want to chop off some of the obviously extraneous bits at this point. While you're at it, rinse the heart under tap water to get rid of blood that might be left in the chambers, check that the aorta is intact, and that there aren't any nicks on the surface of the heart muscle.

The first step in almost all decellularization protocols is to freeze the organ. That's partly for convenience, but the formation of ice crystals will also start breaking open the cells, starting the decellularization. So wrap your bio-hearts in separate plastic bags, chuck them in the freezer, and above all - don't tell your vegetarian wife!

Lots of technical details here - feel free to skip ahead!

This paper has a good overview of the various methods used for tissue decellularization:

An overview of tissue and whole organ decellularization processes

We actually have two different variations of the pig heart protocol: the one originally published by Wainwright et al in their 2010 paper, and the modified protocol from the 2012 video in the Journal of Visualized Experiments, by the same group. Both versions start with freezing at -80C and then thawing the heart, to help break up cells.

Wainwright et al paper:

• Perfuse 3L Type 1 reagent grade water for 15min at 1L/min, then 2xPBS (phosphate buffered saline) for 15min at 1L/min
• Perfuse 0.02% trypsin / 0.05% ethylenediaminetetraacetic acid / 0.05% NaN3 at 37C for 2h at 1 L/min
• Perfuse Type 1 water for 5min, then 2xPBS for 15min
• Perfuse 3% Triton X-100 / 0.05% ethylenediaminetetraacetic acid / 0.05% NaN3 for 2h at 1.3L/min
• Perfuse Type 1 water for 5min, then 2xPBS for 15min
• Perfuse 4% deoxycholic acid solution for 2h at 1.3L/min
• Perfuse Type 1 water for 5min, then 2xPBS for 15min
• Perfuse 0.1% peracetic acid / 4% ethanol for 1h at 1.7L/min
• Perfuse with PBS twice, and Type 1 water three times, 15min each at 1.7L/min

If you add everything together, this protocol takes about 10 hours from start to finish for all the perfusion steps. The modified version from the Jove video goes through much the same steps, lengthening the trypsin and deoxycholic acid treatment to 3h, and the detergent treatment to 2.5h, for a total time of around 13 hours perfusion! They also ramp the perfusion speed from 400ml/min for the initial water perfusion, to 2.2L/min for the deoxycholic and subsequent steps.

In our case, we only had about 10 hours total to accomplish this, and we wound up spending a good amount of that time testing out the pump, fixing leaks in the connections, dissecting down the heart etc. So we wound up adjusting the durations to fit into the time remaining.

On the other hand, since we were making a showpiece rather than a ghost heart that could be successfully reseeded with stem cells, we knew we could easily cheat on some of the ingredients. In particular, we used plain tap water instead of gallons and gallons of expensive "Type 1" reagent grade water. Likewise, we just used plain kitchen salt in tap water instead of phosphate buffered saline. We used OxyClean (sodium percarbonate) instead of peracetic acid, and 151-proof Everclear instead of reagent grade ethanol for the final disinfection step. We also wanted to use SDS rather than Triton X-100 as the detergent, since SDS tends to whiten the tissue better (at the expense of being somewha harsher on the extracellular matrix). We didn't have quite enough SDS, so we wound up going with 2% SDS / 1% Triton X-100. We also left out the sodium azide (NaN3), which is there to inhibit bacterial growth, but is also very toxic - something we'd rather not mess with in a DIY setting.

Oh, and we didn't have access to a high-volume peristaltic pump, but we did have a Bio-Rad Buffer Recirculation Pump (see http://www.bio-rad.com/webroot/web/pdf/lsr/literature/M1702929C.pdf‎) that can theoretically do up to 1.5L/min. Not intended to supply the kind of pressures we need, but meh - let's just see what happens...

So here's what we wound up with in the end:

• Perfuse 3L tapwater for 15min, then high saline (50g kitchen salt in 3L tapwater) for 15min
• Perfuse 0.02% trypsin / 0.05% ethylenediaminetetraacetic acid at 37C for 1.5h
• Perfuse tapwater for 10min, then high saline for 10min
• Perfuse 2% SDS / 1% Triton X-100 for 50min
• Perfuse tapwater for 5min, then high saline for 5min
• Perfuse 2.5% deoxycholic acid (25g in 1L) for 20min
• Perfuse tapwater for 5min, then high saline for 5min
• Perfuse 0.5% OxyClean / 4% ethanol for 1h at 1.7L/min
• Perfuse saline (25g in 3L), then tap water for 5min
• Let drain, then stuff into Ikea jar with 151 proof Everclear!

After freezing, we thawed a heart in the fridge overnight, and took it to BioCurious to set up the perfusion experiment. The "bio-hearts" didn't merely have an intact aorta - they had about a foot of arteries and other gristly stuff attached. So our first task was to dissect all that down.

Luckily, at least one of us had some experience doing dissections... the vegan!

The technique we're using here to push fluids through the heart muscle is called retrograde coronary perfusion. Interestingly, this same technique can also be used in what's called a Langendorff preparation to keep a beating heart alive, by pumping nutrient-rich oxygenated fluids through it. The reason it's called "retrograde" is because you push fluids into the aorta, whereas in normal operation, blood comes out of the aorta. Essentially, we're hacking the cardiac system, taking advantage of a peculiarity of the heart's plumbing! Normally, oxygen rich blood comes from the lungs into the left atrium, from which it is pumped through the aortic valve, into the aorta, and then from there to feed the rest of the body. But the very first thing that gets fed by this freshly oxygenated blood is the heart muscle itself, through two coronary arteries located just above the aortic valve (see the schematic above). Now, when we push fluids retrograde into the aorta, that pressure will cause the aortic valve to close, and the only place for that fluid to go is through the coronary arteries into the heart muscle - cool, no?

After separating the aorta cleanly from the pulmonary artery and trimming it down a bit (to just below the brachiocephalic artery, the smaller artery connecting to the aorta closest to the heart), we inserted a brass barbed hose connector into the aorta (5/8" into the aorta, to 1/2" for the tubing) and tightened it securely with a wide hose clamp. But apparently not tightly enough, since we accidentally pulled it out a couple steps later...

After connecting the 1/2" vinyl tubing to the heart and the pump and tightening all the connectors with hose clamps, we lowered the heart into 3 liters of tap water in a giant 4 liter beaker, turned on the pump - and discovered we had sprung a leak! Apparently our butcher wasn't too careful and put a tiny nick in one of the small arteries on the surface of the heart, through which a lot of our pressurized water was now escaping...

Remembering that superglue (cyanoacrylate) has been used to replace stitches in surgery, we decide to give it a try on our little leak. Worked perfectly!

• Perfuse 3L tapwater for 15min, then high saline (50g kitchen salt in 3L tapwater) for 15min

We start out the perfusion with a round of water, and a round of high saline. The first couple of perfusions will rinse out a lot of residual blood that is trapped inside the heart muscle. The extreme change in salinity will also cause the cells to swell and burst by osmotic pressure.

If you want to take some gruesome pictures, this is your photo opportunity, as streamers of blood start pouring out of the heart! Note how the blood will stream down in plain water, but it will float up in the saline, due to the difference in density.

We will do another short round of water and then saline after each of the subsequent steps as well...
 

• Perfuse 0.02% trypsin / 0.05% ethylenediaminetetraacetic acid (EDTA) at 37C for 1.5h
• Perfuse tapwater for 10min, then high saline for 10min

Trypsin is a protease, that is, an enzyme that breaks down other proteins. Trypsin is most active at body temperature, so warm the solution to 37C in advance. EDTA acts synergistically with trypsin: it is a calcium chelator, which integrins needs to interact with other proteins for cell adhesion, so EDTA by itself will loosen the bonds between cells. But it also acts to mop up divalent cations that would otherwise inhibit trypsin activity.

This step is one of the workhorses in terms of breaking open cells, and degrading and flushing out their content. You'll still see a lot of blood coming out, but it's definitely not just blood alone, and by the end of this step you'll notice the heart getting significantly paler.

Trypsin can be purchased online but If you can’t get your hands on trypsin or want to do a DIY version, you could use other proteases that are much easier to find. Bromelain is the enzyme that is responsible for making the inside of your mouth hurt after eating too much raw pineapple. Fresh papaya has a similar enzyme called papain. Both of these are fairly easy to find as digestive enzymes at your local health food store. Or in your local grocery store under "meat tenderizer"!

These are fairly gentle digesters but if the heart is left in them for excessively long, the collagen scaffolding can also be degraded. The protocol from the Jove video recomends 3 hours of trypsin perfusion, and you probably shouldn't go any longer than that.

• Perfuse 2% SDS / 1% Triton X-100 for 50min
• Perfuse tap water for 5min, then high saline for 5min

Detergents work by breaking up fats, and since cell membranes are essentially made up of fats, detergents can be used to bust open cells very effectively. SDS (sodium dodecyl sulfate, also known sodium lauryl sulfate) is a commercial detergent that is often used in the laboratory. It also happens to be one of the most common detergents found in shampoos, so for a hard-core DIY version, you could probably use a strong shampoo for this step! If you do use Triton X-100, as the published protocol recommends, it it useful to make this up about 45 mins or more before you will need it as it can take time to get the Triton X into solution (we used a magnetic stirring plate and patience).

By now you'll notice that all the blood has been removed from the heart, and what is coming out is brownish cellular content...

• Perfuse 2.5% deoxycholic acid (25g in 1L) for 20min
• Perfuse tapwater for 5min, then high saline for 5min

Deoxycholic acid is what's known as a secondary bile acid, meaning that it is actually produced by your gut bacteria from a primary bile acid secreted by your liver. In the body, deoxycholic acid aids in digestion of fats by emulsifying them.

Here, deoxycholic acid is used to break up and flush out more fats from the tissue, similar to the earlier detergent step. Since this is one of the more expensive ingredients, you can probably get by with doing another round of detergent perfusion if you're just making a show piece.

• Perfuse 0.5% OxiClean / 4% ethanol for 1h at 1.7L/min
• Perfuse saline (25g in 3L), then tap water for 5min

This is the final step where the heart will be flushed with a solution of oxyclean and ethanol to disinfect any bacteria that managed to stick around throughout this process, and flush out any last cells from the previous procedures. The official protocol uses peracetic acid, a strong oxidizer that you could produce yourself from vinegar and hydrogen peroxide. OxiClean works great as an oxidizer though, so we used that instead.

Once the disinfectant step is completed, and you have completed the final saline rinse you can remove the heart from the tubes. Let it drain - we used a plastic box lined with paper towels. You'll notice that the Ghost Heart is quite floppy at this stage, and will deflate like a baloon as water drains out.

We purchased a 1/2 gallon Ikea Korken jar with lid. During the course of decellularization, the heart swells to about double its original volume, and we were concerned that getting it through the neck of the jar might be a tight fit. Turns out the Ghost Heart is floppy enough at this stage that it will easily fit into the jar.

We used two 750ml bottles of Everclear total for the disinfection step and bottling - one of the more pricey items in the procedure.

For visual appeal, and to help us suspend the heart in the jar we opted to leaved the clamps and brass fitting attached. Be very careful as the heart will be somewhat delicate at this stage. We were then able to put it into the jar and fill it with the 151 Everclear. The heart will remain on the bottom if you dont suspend it. We used two rubber bands attached to the brass fittings and hooked around the lid to give the heart the appearance of being suspended in the jar.

Clean and disinfect your equipment and surfaces.

Enjoy your showpiece!

We took our Ghost Heart to the "Creatures of the NightLife" Halloween event at the California Academy of Sciences, a big over-21-only mad science Halloween geek-fest. Big hit! We even had the full decellularization setup running with a second pig heart and giant beakers filling up with pig blood. We also had a preserved pig heart from Carolina that we could let people hold - with gloves obviously. (Just $8.75 for a preserved pig heart? What a steal!) Check out some more pics from our Halloween show on our meetup page - a good gruesome time was had by all!