Hi Sharoon Akhtar

How are you purifying, and what is causing you to conclude your protein is oligomeric? Why so much glycerol? If you can provide some more information about the target that might help diagnose the problem. Some proteins may natively oligomerize and this may be required for their function, so disruption of the oligomer could result in protein that is of no use. Is there a reason you need the protein monomeric?

Some things to try if you haven't already:

-Analytical gel filtration or SECMALS to assess degree of olgiomerization and if any protein remains monomeric. You may need to repeat this after attempting any changes to your purification/storage conditions to see the effect

-Reduce your protein with BME/DTT/TCEP to check if disulfides have anything to do with the oligomeric state. Some complexes can be held together by intermolecular disulfide bonds

-Reduce the glycerol concentration

-Change the ionic strength of your purification/storage conditions. Usually more salt would be better but this is protein specific

-Remove the SUMO tag - Ulp1 protease is commonly used for this

-If all else fails, sometimes sub-denaturing concentrations of urea/guanidine will disrupt an otherwise very stable complex. But, its removal may lead to the complex forming again.

Worth mentioning is that N-linked glycosylation will not be happening in an E. coli host so this is not a factor.

Best of luck,

ACA

Hi Alexander C. Anderson ,

Thank you for your input and some helpful insights. Here is some more information.

Lysis/Binding buffer: Lysozyme 1mg/ml, NaCl 300mM, Tris 100mM, and Imidazole 5mM, 20% Glycerol | Elution Buffer: NaCl 300mM, Tris 100mM, and Imidazole 500mM, 20% Glycerol. I am not adding glycerol in the wash buffer with 70 mM Imidazole.

Observation: There was less oligomerization in the lysate from one experiment. However, it increases in the Flowthrough of Ni-NTA | Hypothesis: Perhaps the protein requires a stabilization molecule to keep it in a monomeric state.

Why so much glycerol? I tested 10% glycerol from Article Optimised production of disulfide-bonded fungal effectors in...

I removed the tag once and ran on SEC; the results were inconclusive. I am planning to do it once more to see if it helps.

I agree that some proteins are stable in oligomeric forms. However, in another SEC experiment, there is also a small fraction of monomeric isoforms. I am also interested in the monomeric forms of this protein for crystallization experiments.

You mentioned reducing agents. Do you have experience with reducing agents, and what is the minimum concentration to start with without disrupting the disulfide bonds? I don't have access to SEC-MALS but I can do DLS to identify molecular weights in the fraction.

N-glycosylations also play an essential role in solubility; I wanted to know if, in E. coli the absence of these glycosylations may have to do something with the oligomerization.

Please let me know if this information help you identify the cause of oligomerization or alternative strategy to get monomeric forms.

Hi Sharoon Akhtar

Ok, great. The first thing I notice is that your protein experiences a 0.9M ionic strength in your Ni-NTA elution, which is very high. Some proteins may tolerate this, but you could reduce or heavily limit the amount of NaCl in the elution buffer and see what effect this has. Alternatively, you can dialyze your elution against a lower total ionic strength buffer. This is highly protein specific, but in my (anecdotal) experience most proteins tend to be most stable at 0.1-0.3M.

Normally, I would ask how concentrated this enzyme is from the column and when storing it, as some proteins will have a propensity to oligomerize only above a certain threshold concentration. But, you've mentioned that you are aiming to crystallize this, and so concentrating this protein well above its cellular concentration will be more or less necessary. Is there a reason you are preferring the monomer for crystal screening? I don't see a reason to avoid screening with oligomeric protein unless you have evidence that the oligomer results in an extreme unit cell dimension or is otherwise resisting crystallization in some way. Actually, one might argue sampling for a condition that causes the protein not to self-associate would be counter-productive for crystallization - after all, you're trying to create a crystal, which is just a very large array of ordered, self-interacting protein.

In terms of disulfide bonds, most protocols will intentionally reduce the protein completely with an excess of reducing agents, usually 1 mM. You could, in theory, multiply the number of disulfide bond pairs of your protein with its effective concentration and titrate in the reducing agent to see what effect it has. A simpler method might be running an SDS-PAGE without reducing agent in the loading buffer. If your protein first runs at the expected size of the monomer, but then oligomerizes when run with a nonreducing loading buffer, you can infer that a disulfide bond is implicated.

It would be difficult to say what effect glycosylation might have. I can't think of any specific examples where a glycoprotein sampled an oligomeric state it would not otherwise when expressed glycan-free, but I have no experience there. My intuition there would be that I doubt it plays any role.

ACA

Thank you Alexander C. Anderson ! This is very helpful. I can give it a try and see how it goes.