A couple of tips: 1) If the protein is that large, you should be using a C4 stationary phase with much wider pores - as you suggested (Something like this: https://www.perkinelmer.com/product/supra-clean-spe-c4-200-mg-3-ml-pkg-50-n9306593 ) 2) Make sure you aren't overloading the SPE cartridge with too much material relative to the sorbent bed. The manufacturer should have some guidance on this. 3) Double check that the pH of your loading buffer is around 3 or less. Try to get the pH close to your equilibration buffer ( 5% ACN, 0.08% Trifluoroacetic acid). You have alot of imidazole, which is likely buffering the solution quite strongly. 4) MWCO filtration could also work here (depending on the amount of protein you have) - you could buffer exchange into something like 25% ACN, 0.1% TFA before lyophilization.

Hello, use dialize technique to reduce the salt content in sample.

Hey James M Fulcher!

Thank you for your helpful answer!

If you don't mind me asking:

1) I agree with the pores, but can you explain why C4?

2) I load 6mg protein (5% of resin) on 200mg resin. I think that should be fine when it comes to capacity.

3) Good point. I will try and check that my sample is close to the pH of my equilibration buffer. This one I described is the only one of those three that has such high imidazole content. The others are in 350mM Imidazone, but I guess I should be careful with the pH there as well. However, even if the pH is a bit higher (let's assume around 6-7) then I would still expect them still to bind. Or am I missing something?

4) The point why we are not using MWCO filtration is due to the fact that our proteins tend to start aggregating at some point. We just don't want to risk aggregation. However, if I keep the concentration low by keeping the volume up, it might be possible.

Thanks for the answer Karen A. Darbinyan !

Could you please elaborate on why I should reduce the salt concentration in my sample? Do you suggest to dialize into lower salt and then binding should work?

The ionic strength of the solution, the size, valence, and nature of the salt ions affect the thermodynamic as well as the mass transfer kinetics of the adsorption mechanism of C18 column.

See link below for more info.

https://www.researchgate.net/publication/8630071_Effect_of_the_ionic_strength_of_salts_on_retention_and_overloading_behavior_of_ionizable_compounds_in_reversed-phase_liquid_chromatography_I_XTerra-C-18

Good luck!

Thanks for this information!

Right now I am talking to several people and companies for column recommendations and will also follow James M Fulcher tip to order and use another column with suitable particle and pore size. I will also try to reduce ionic strength and I will check the pH of my sample now.

The column pore size you selected is much prone to perform peptide level analysis rather than larger biomolecules.

If you choose C4 ligands nothing will change since you are mentioning you are observing flow through even the usage of more hydrophobic C18. C4 and C8 are typical ligands for large protein targets. Your exclusion limit resulting from the chosen pore size is hindering the interaction between your protein and bonded phase. Thus you should increase pore size to at least 300A.

By the way, the high amount of salt induces hydrophobic interaction, you may remember from the HIC applications. Desalting does not improve retention.

If you need to narrow down the problem you can inject well-retentive protein such as BSA and make a query whether your chromatographic approach is suitable or not for protein analysis.

Update: The pH of my washing/equilibration buffer is 2.2 and my sample pH is about 7. I hope that this wouldn't be too much of an issue to be honest? The isoelectric point of all my recombinant proteins is above 9.

Additionally I attach a picture of a gel from the run:

From left to right: Input/Marker/flowthrough1/flowthrough2/20%Elution/40%Elution/60%Elution/80%Elution/spilover/marker

Percentage in elution refers to the ACN amount.

İsmail Emir Akyildiz

Thank you!

Yes, as far as I know, high salt increases hydrophobic interactions. So the salt should not influence my retention that much. Thank you for reminding me.

I talked to the support of Phenomenex(R) and will receive some columns from this company to try them out. One of with has 260A and the other 300A pore size. Both of them will be organic polymer-based resins.

Lets hope they will work :)

Hi Philipp Czermak , good questions.

With regards to C4 vs C18 - I'm speaking from personal experience and the recommendations of chromatographic companies. Alkyl chains longer than C4 tend to produce more peak tailing/broadening and lower recovery of large peptides/proteins. This is what I've seen with HPLC separations. I don't have a great theoretical answer as to why shorter alkyl chains are advantageous, though there are many potential explanations that could be put forth. With pore size, consider that the smaller pores limit the ability of larger proteins to diffuse onto the stationary phase. Large proteins diffuse quite slowly relative to smaller molecules, and this behavior is only exacerbated with smaller pore sizes.

With regards to pH, there is another potential scenario that relates to the small molecules in your sample buffer. With imidazole, if your pH is around 7 that means around half of the molecules are neutral at any given moment. It is possible, though not all that likely, that uncharged imidazole (or DTT or HEPES) could bind to the C18 stationary phase - meaning you are oversaturating the sorbent bed with smaller molecules that are present at much larger amounts. Shorter alkyl chains such as C4 can also help in this case, as the retention of smaller analytes is very low with short alkyl chains.

Typically, it always best practice to match the pH of your sample to your starting mobile phase. If you do not, you risk the protein precipitating before it even has time to bind to the stationary phase.

Thanks for the advice! I appreciate it. However, with that much Imidazole in my buffer, it is hard to lower the pH even with 0.08% TFA. And the maximum TFA used is usually 0.1%. Therefore I wont get much lower than pH 7. However, luckily it seems that my protein is not precipitating or crashing on the column due to the sudden pH difference.

For now, I will wait for the columns with bigger pore size since my best guess is the pore size. And maybe I try to find someone with an TSKgel TMS C1 column since this is the only specific column I found that the group was using that we try to replicate the results from.

Why not do a buffer exchange on a sephadex column instead? Or at least as a preparation for the RPC to get rid of the imidazole.

Bo Pontoppidan the sephadex I’m not using since I think it will destroy my column due to the ACN if that’s what you mean? Otherwise dialysis, SEC or desalting is a valid option, I agree. But for now, my problem is the pore size that is too smal and I would like to test step by step. If the new columns will give me unsatisfying results, then I will definitely try bufferexchange.

Hey guys!

Just a quick update! So, I got those two columns from Phenomenex (Strata-XL and SDB-L). I decided to go with the same protocol as before for the first trial. The only differences were the new columns and that I used an adaptor for a syringe to apply some pressure to increase flow rate (mainly during conditioning, washing and elution).

Results: Both columns seem to work better than the one I used before (C18 with 125A pore size). However, the Strata-XL (100um diameter, 200mg resin and 300A pore size) seems to not capture all my protein, whereas the SDB-L looks good and just shows minimal breakthrough. Attached I send also a picture of one of the three proteins I am working with. The other two look fairly similar.

I'm really happy and excited. Proteins are already lyophilizing in the freeze-dryer ;)

Thanks a lot for your help!

Cheers

Bo Pontoppidan Since without either Imidazole or GndHCl my protein would aggregate.