Problem with gradient HIC

[HW Mueller]

Dancho, some of these mismatches were done under extreme conditions to learn more on these phenomena. Thus some mobile phases were very low concentrations of NaOH and H2SO4. Even 0.5 µL injections could give exreme results. However, I was also surprised by some "misbehavior" with more "normal" mobile phases and inj. volumes around 5µL.

qrys, injecting the protein in 0.8M salt didn´t improve things? If this is correct, and also the protein is out of the column before the salt concentration reaches 0.2M, one runs out of explanations. Thus I also tend to go in the direction of questioning the column "chemistry".
I have precipitated proteins (mostly albumin) with PEG, also how stable is this PEG coating?

[qrys]

Danko, I made these columns in our lab, the idea was to study the effect of hydrophobicity of surfaces on protein separation. So I have different batches of silica particles silanized with PEG silane (actually 3 EG segment) and alkyl silane. I started with pure PEG surface as it's the most hydrophilic and will give the least effect on denaturating protein upon interaction with the surface.

HW, protein in 0.8 M salt didnt improve things. I think the silane bonding on the silica is pretty stable as I have used these columns with repeat injections of sample isocratically and the results are consistent.

Uwe, there are some old papers (1980s) studying EG coated silica particles for HIC application, for examples the work of Horvath and Karger. Lysozyme was one of their test proteins.

My advisor (who is not a chromatographer) thinks that the problem is technical ie. related to the system, he thought the chromatogram looks like I ran the gradient in reverse (i.e. no salt to high salt), which would explain the wide peak. However, I ran the right gradient, I checked all the lines etc. I monitored the flow from solvent A and B. Unless there is something wrong with the gradient mixer?

[Uwe Neue]

Tell your supervisor to by a textbook on HPLC. I recommend Uwe Neue, HPLC Columns. In Chapter 13, there are some HIC gradient chromatograms, which show an increase in peak width with retention in a salt gradient. Also, you can find there some equations that deal with the retention mechanism.

The theory looks very straightforward, and it should be a no-brainer to solve the equation for the retention factor at the point of elution and then the peak width. I do not have the time now to do so, but it should be easy.

[danko]

If I were in this situation I’d buy a commercially available HIC column (e.g. phenyl, propyl or something like that) with a proven performance (there are lots and lots of brands out there), obtain the desired chromatography utilizing that column and applying conditions suggested earlier on this thread and then go back to the “home madeâ€

[qrys]

Danko,

I pack the particles into waters glass column, dimension 5 mm x 5.5 cm.

I'll try adding the IPA, instead of IPA can I use ethanol? I wonder why IPA is very widely used in HPLC work and why not ethanol.

If I have isocratic data for lysozyme on this column, should I be able to guess what is happening for gradient mode?

Uwe, I'll figure out the calculation.

Thank you

[Uwe Neue]

The theory equations form Horvath should permit a fit to the retention data.

[danko]

Qrys,

With the given dimensions and flow rate of 1 mL/min I would call the peaks you mentioned earlier â€

[HW Mueller]

qris, I can´t figure out what gradient gave you the peak with baseline of 30 min, nor do I see whether you injected in water. Not having Uwe´s book in front of me, nor Horvath´s theory, I can´t imagine what sort of calcs could give such results. Right now it looks like the protein precipitates, then very slowly redissolves.
Did you ever vary, the amount of protein injected, strongly downward?

[qrys]

HW, I never vary the concentration of the protein.

If I cant dissolve the protein in 1 M Na2SO4, then it should precipitate in the column when I started the gradient. Is protein precipitation in column during HIC normal? if it is not, then it may be the problem of protein redissolving giving wide peak?

Danko, I'll post the chromatograms soon...

Thank you

[qrys]

Hi, I have 4 figures here:

1) isocratic data of lysozyme eluted at various concentration of Na2SO4


2) chromatogram of lysozyme elution from 50 min gradient of 1 M to 0.5 M Na2SO4, followed by 5 min gradient from 0.5 M to 0 M of Na2SO4. Mobile phase A: 1 M Na2SO4 in phosphate buffer pH 7, B: phosphate buffer pH 7. Sample: lysozyme 5.5 mg/ml in water, 5 ul injection volume.



3) chromatogram of lysozyme elution from 50 min gradient of 1 M to 0.5 M Na2SO4. Mobile phase A: 1 M Na2SO4 in phosphate buffer pH 7 and 10% IPA pH 7, B: phosphate buffer pH 7 with 10% IPA. The chromatogram is not complete 50 min because the lysozyme was eluted very early.
Sample: lysozyme 5.5 mg/ml in water, 5 ul injection volume.



It seems with IPA the sample is almost like unretained?

4) chromatogram of lysozyme elution from 50 min gradient of 1 M to 0.5 M Na2SO4. Mobile phase A: 1 M Na2SO4 in phosphate buffer pH 7 (no IPA). B: phosphate buffer pH 7 with 10% IPA>, this is then followed by isocratic 0.2 M Na2SO4 to remove lysozyme from the column.






[/img]

Could you explain me the role of IPA here?

Thank you very much, I appreciate you guys are very patient with me

[danko]

I see some progress here

Next step: Try a steeper gradient f. x. 1.5 and maybe 2 % per min.
Keep the eluent A IPA-free, while reducing the IPA content in eluent B to 5 – 7 %.
And now to your question:

Could you explain me the role of IPA here?

The high concentration of Sodium Sulphate in eluent A (initial condition) facilitates structure making in the bulk water, which removes the structured water around the protein’s hydrophobic amino acids, which in turn exposes these hydrophobic entities to the relatively hydrophobic stationary phase causing mutual attraction.
When the gradient starts the salt gets diluted which results in less and less structured water, which in turn allows for water structuring around the hydrophobic amino acids and that makes the protein more water soluble (i.e. prefers the mobile phase over the stationary ditto).
IPA (which is a chaotropic/structure braking agent) just adds more chaos to the environment thus facilitating the protein’s solubility in the mobile phase, which in turn speeds up its elution.

Back to the experimental work. I think there are good chances of chromatography improvements if you follow the path above. But I’m not completely convinced that it’ll ever get perfect using PEG as a stationary phase. Maybe it doesn’t matter in this case as your work is aiming at showing some tendencies and not necessarily at demonstrating an excellent chromatography.

Best Regards

[qrys]

Dancho,

Here is the chromatogram of lysozyme elution with 1-0.5 M Na2SO4 for 25 min. A: 1 M Na2SO4 in phosphate buffer pH 7, B: phosphate buffer pH 7 with 5 % IPA. Flow rate 1.5 ml/min. It seems that the protein recovery was increased as compared to the previous result (figure 4 previous posting).



You mentioned earlier that PEG phase may not be good for HIC, what kind of phase would you suggest for HIC? What I see on papers is typically the sepharose series, with butyl or phenyl ligand. Arent they similar to PEG coated silica in terms of the surface hydrophobicity towards protein?

Thank you

[danko]

Yes, Sepharose matrix is one possibility and there are others.
Regarding PEG hydrophobicity compared to Butyl and Phenyl, I’m not completely sure, but in my mind PEG is less hydrophobic than these two options. Another important thing to keep in mind is the matrix itself (the stationary phase support). While your support is silica, the Sepharose is a crosslinked agarose, which is quite hydrophilic.
Besides, especially in HIC the degree of derivatization is quite important. In order to achieve reasonable HIC, it’s important to avoid multipoint interactions between the protein (analyte) and the stationary phase ligand. Multipoint interactions can result in practically irreversible retention of some species. And then there is the pore size, the particle size, distribution etc.
So, as you see there are many parameters that influence both the selectivity and the efficiency.
Taking into account the materials and experience you have, I think the latest experiment is about the best achievable result.
If you’re planning more comprehensive work involving HIC, I’d still suggest that you bought a small amount commercially available stationary phase (could be butyl derivatized sepharose) packed a column with that and used it as a reliable reference.

Best Regards