How do I Get my Protein Out of Inclusion Bodies?

Have you ever had a problem where you had to try many different potential solutions before you finally found one that worked? Early in my doctoral research, I had a problem like this. I needed to get my full-length protein in sufficient quantities and excellent purity to enable a certain kind of experiment.

I made dozens of attempts to optimize the expression conditions and the purification protocol. Some strategies worked a little better, some a little worse. Yet, even with all of these attempts I still seemed to be miles away from my desired result. Cue the sad trombone!

Finally, I found a solution that gave me gobs of protein with much better purity than anything else I had tried. What worked? I ramped up the expression conditions to drive the protein into inclusion bodies. My precious protein was protected from degrading proteases while inside the inclusion bodies, and after getting rid of those pesky proteases, I then resolubilized my protein of interest out of the inclusion bodies, and refolded them back into an active conformation.

Proteins are extracted from inclusion bodies using solubilizing salts like guanidine and reducing agents like TCEP. The extracted proteins are then purified and refolded. Functional or structural analyses should be performed to verify that the protein is refolded into an active state.

In this article we’ll discuss what I learned about refolding my protein of interest from inclusion bodies and what you should keep in mind if you find your protein of interest stuck in inclusion bodies as well.

In this article:

What are inclusion bodies?

Inclusion bodies limit toxicity

Inclusion bodies prevent proteolytic cleavage of the expressed protein

Breaking your protein out of inclusion bodies

Making sure your protein is still functional

Related Products

References

What are inclusion bodies?

Inclusion bodies are insoluble pellets of protein that form when a protein is overexpressed in a host cell (Figure 1). Inclusion bodies are particularly common in E. coli and other bacteria that have a strong reducing potential compared to eukaryotic cells (Gąciarz et al, 2017; Singh et al, 2015). This means that disulfide bonds between cysteine amino acids in different proteins are more prone to form in E. coli, which contributes to inclusion body formation (Figure 1). It’s estimated that many proteins will form inclusion bodies in E. coli when expressed at a level higher than 2% relative to the native bacterial proteins (Mitraki et al, 1991).

Figure 1. Inclusion bodies form at bacterial poles (left) and consist of proteins (purple) clumped together into an insoluble mass (center) that is often aided by intramolecular disulfide bonding through cysteine residues (right).

So, what do you do if your protein mostly expresses into inclusion bodies? Usually, people try to avoid inclusion bodies. By dialing down the expression conditions, or by adding a solubility tag, you can often prevent your protein from going into inclusion bodies in the first place.

But for some proteins, inclusion bodies may either be unavoidable, or may even provide certain advantages for your purification. What advantages can come from this sticky mess? I’m so glad you asked.

Inclusion bodies limit toxicity

Many proteins are toxic to the host organisms within which they are overexpressed. This was likely the case for the protein I was trying to express during my PhD research.

It was a human transcription factor, so when expressed in E. coli, it would have bound all over the bacterial genome and disrupted the normal transcriptional programs that enable bacteria to grow and thrive. As a reminder, transcription is the process of transcribing DNA into RNA messages, a ubiquitous process that takes place in all lifeforms. Therefore, cells that expressed my protein highly would have stalled during culture, leaving only low-expressing cells to replicate and thrive. You can see how this limits the total amount of protein I would eventually get out in the end.

Ultimately, I found a protocol that limited toxicity by preventing early “leaky” expression of my protein of interest before the cells were ready, and then rapidly expressed my protein forcing them into inclusion bodies (Studier, 2005). Once inside the inclusion bodies, my protein was sequestered from the bacterial genome and was unable to disrupt the normal bacterial transcription. This meant the bacteria kept growing just fine and producing more of my precious protein.

Inclusion bodies prevent proteolytic cleavage of the expressed protein

The other issue I was having with previous expression conditions is that my soluble protein was getting cut up by bacterial proteases during expression. This protein has lots of disordered regions which are easy for proteases to chop up. Protease inhibitors can be added to prevent protein degradation during the purification process, but my protein was getting chopped up even during the expression stage. So, at the end of my purification I had some of my desired protein, but I also had lots of degradation products that contained only parts of the protein I wanted (Figure 2).

Figure 2. Proteases (scissors) cut proteins when they are in the soluble fraction (left). But when the protein is in an inclusion body, it is inaccessible to the proteases (right).

Luckily, when my protein was expressed into inclusion bodies, the proteases were no longer able to cut it up. The inclusion bodies physically sequestered the protein away from deleterious proteases, meaning that my protein wasn’t getting chopped up anymore (Figure 2). As you’ll see in the next section, I washed the proteases away during the purification before releasing my protein from the inclusion bodies. So, once my protein was soluble again, there were no proteases left to cut them up.

How to break your protein out of inclusion bodies

Getting your proteins out of inclusion bodies the make-or-break step. It doesn’t matter how much protein you have in those inclusion bodies if you can’t resolubilize and convert them into an active format.

First, you want to take advantage of your protein of interest being in inclusion bodies as a purification step. That is, you can use your protein’s isolation in inclusion bodies to your advantage in separating it from other soluble proteins.

After lysing your cells, you centrifuge the lysate to separate the insoluble inclusion bodies from the soluble proteins (Figure 3). Then, wash the inclusion body pellet a few times with your lysis buffer. By this, I just mean to add buffer to the pellet, physically resuspend the pellet using sonication, pipetting, or stirring, and then centrifuge the solution to spin down the inclusion bodies again. This will ensure that you get rid of all soluble proteins, such as those pesky proteases I mentioned above, before breaking your protein of interest out of the inclusion bodies.

Figure 3. Centrifuging cell lysate leads to inclusion bodies (purple proteins) in the pellet whereas other proteins (green and orange polygons), including proteases (scissors), remain in solution (left). The soluble fraction is pipetted into another tube to separate the inclusion body from the rest of the proteins (right).

Now it’s time to free your protein of interest from its insoluble isolation. In this article, we discussed the Hofmeister Series of salts and how some salts “salt in” proteins whereas other salts “salt out” proteins. Urea and guanidine are two salts that very strongly salt in, or solubilize proteins.

People usually use 6-8M urea or 6M guanidine to solubilize proteins out of inclusion bodies. Additionally, you’ll probably need a reducing agent like DTT or TCEP to break the disulfide bonds we mentioned above that are trapping your protein in inclusion bodies. So, you’ll prepare your lysis buffer with guanidine and TCEP, for example, physically resuspend the inclusion bodies one more time, then incubate for around 30 minutes to an hour to let these salts do their solubilizing work.

This time, when you centrifuge once more, much of your protein will be in solution instead of in the pellet at the bottom of the tube. Often, an initial purification step can be conducted with urea or guanidine in the buffers. His-tags and nickel (Ni²⁺) columns work just as well with these harsh salts, and so you can perform a purification step to isolate your protein from any other proteins that were also in the inclusion bodies.

So, you now have your protein solubilized and partially purified - yay! But here comes the hard part: while urea and guanidine are really good at salting in most proteins, they are also destabilizing to the protein’s structure. This means that your protein will be solubilized and partially purified, but it will also be denatured. It lacks the native fold that enables its function (Figure 4). You’ll now need to remove the urea or guanidine in a manner that allows the protein to regain its native fold, without allowing it to aggregate.

Figure 4. Proteins have a native structure (left) and a denatured state which is devoid of any secondary or tertiary structure (right). While solubilizing proteins out of inclusion bodies, salts like guanidine and urea push proteins into the denatured state.

Oftentimes, people will use dialysis to dilute out the urea or guanidine. While this sometimes works well, it also “refolds” the protein quite slowly, over the course of several hours while the denaturing salt is dialyzing away. This will sometimes allow sticky protein intermediates sufficient time to bind to one another and aggregate (Figure 5)(Yamaguchi and Miyazaki, 2014).

Figure 5. Partially

refolded protein states are often sticky and can bind to one another during

dialysis leading to protein aggregation and loss of your protein.

Dialysis is a fine approach to try first, and can be optimized (Modanloo Jouybari et al, 2018). But if you’re having difficulties with dialysis, I’ve found that refolding on the column often works better for many proteins. That is, while performing the Ni²⁺ purification of your his-tagged protein, switch to a buffer without urea or guanidine while your protein is still bound to the Ni²⁺ column. This allows the refolding to occur quickly, in a matter of seconds, avoiding prolonged periods of sticky, intermediate states. Also, the individual proteins are bound to the column, so they lack the freedom of motion necessary to bind to other proteins to aggregate (Singh et al, 2015). After “refolding” your protein on the column, you can now elute your refolded protein off of the column just as you normally would. If you want more details about his-tagged proteins and Ni²⁺ columns – see this article.

Not all proteins can be successfully refolded from a denatured state. In general, single domain proteins can usually be refolded whereas multidomain proteins with complex quaternary structure are more challenging (Kerner et al, 2005). Finding the optimal refolding conditions can require optimizing multiple steps, but if after many attempts you’re still not getting any soluble protein, you may want to revisit the expression conditions or try adding a solubility tag, as we discussed above.

Making sure your protein is still functional

For the sake of this article, let’s assume you successfully liberated your protein out of inclusion bodies, and it is still in the soluble fraction after getting rid of urea or guanidine. Congratulations – that may have taken a lot of work! Now, the last step is to make sure that your protein is functional, or active.

How exactly you do that depends on what type of protein you’re working with. The protein I was purifying was a transcription factor that binds to DNA. So, I used the protein to perform quantitative DNA-binding assays and demonstrated that the protein refolded from inclusion bodies bound to DNA just as tightly as protein that was expressed in the soluble fraction (Figure 6). This suggested that I had correctly refolded my protein into an active conformation.

Figure 6. The protein I was working with was a DNA-binding transcription factor, so I compared how tightly the protein bound to a specific DNA sequence when it was expressed in the soluble fraction (left) or recovered from inclusion bodies (right). Free DNA (lower gray bands) migrates quickly through the native PAGE gel, whereas DNA bound by the protein migrates slower (upper gray bands). The transition from free to bound DNA happens at the same protein concentration, indicating that protein purified either way has the same DNA-binding activity. See Currie et al, 2017 in References for more details.

Additionally, collaborators and I used this protein to determine the structure of the DNA-binding domain (Currie et al, 2017). This structure looked identical to one solved by another lab using protein expressed in the soluble fraction, within expected experimental error (Cooper et al, 2015). Solving the structure to make sure that you refolded the protein correctly is probably a little overkill for most projects, but you definitely want to benchmark your protein with some kind of enzymatic or binding assay to make sure it is functioning as it should.

If you’re working with proteins that love to go into inclusion bodies, this is how you could attempt to free them from their insoluble prison and release them out into solution. Keep in mind this procedure doesn’t work for all proteins, so it’s usually worth trying to optimize expression conditions or using a solubility tag as well. But when purifying proteins out of inclusion bodies does work, there can be added benefits such as enhanced protein yields and superior purity – as was the case for my protein.