Peak width and injection volume- proteins/peptides

[pipettemonkey]

I understand small molecules and proteins behave differently on the reverse-phase HPLC column.

I'm told proteins have one adsorption and one de-desorption step during the HPLC run (Vydac manual). Small molecules continually adsorb/desorb, and peptides are said to be a hybrid of both.

It's easy to see why larger injection volumes give broader peaks for small molecules, but what about proteins? Peptides?

I'm isolating a peptide. I expect it to be between 30-50 amino acids. I've been getting very broad regions of activity on all columns (C4, C18, polymer). I've also been using large injection volumes (and multiple injections).

Is this why my region of activity is so broad?

[Uwe Neue]

There is not an on-off difference between small molecules and proteins. The difference is gradual. The one adsorption/one desorption step for proteins is a drastic oversimplification, which can lead you astray, if you believe it to be literally true.

The real isue is the steepness of a plot of the natural logarithm of the retention factor versus the solvent composition. Small molecules have values around 10. Peptides have values around 20 to 40, and proteins have values around 80 or higher. 80 is 8 times steeper than 10, but it is not "on-off". A large injection volume for a peptide will also result in a larger elution volume. Not as bad as if this would have been for a small molecule, but you will still see measurable effects.

[danko]

Just to add a bit to Uwe’s excellent explanation: Large injection volume is not just a large volume. Things can get worse if the sample solvent temporary disturbs/changes the pH of the eluent (when important and it typically is important in protein context) or f. ex. dilutes the eluent in terms of salt concentration (again very important factor in protein context). And then more exotic effects could be temperature disturbance (if the protein solution is very cold), too long injection process etc.

Best Regards

[pipettemonkey]

Just to add a bit to Uwe’s excellent explanation: Large injection volume is not just a large volume. Things can get worse if the sample solvent temporary disturbs/changes the pH of the eluent (when important and it typically is important in protein context) or f. ex. dilutes the eluent in terms of salt concentration (again very important factor in protein context). And then more exotic effects could be temperature disturbance (if the protein solution is very cold), too long injection process etc.

Best Regards

This is a good point.

I am purifying natural ligands. This means 6-8 HPLC steps. I typically run CH3CN + 0.1% TFA in the mobile phase for all HPLC steps (I have used 1-propanol, but always stick with the TFA or HFBA as ion-pairing agent).

Instead of drying down the fractions from the previous run, I just dilute them with 2-3 volumes 0.1% TFA and re-inject. This makes for large injection volumes that still contain 5-10% CH3CN

This is too low organic % to elute my ligand, however, perhaps it is enough to spread the band down the column while I am injecting the rest..?

The reason I don't dry the fractions before proceeding to the next step is fear of losing some precious purified ligand to irreversible binding to the PP tube upon drying.

I'm gong to add 100 micrograms of inert protein, such as BSA, to my tubes and dry them down. Hopefully when I bring the residue back up in a small volume of starting mobile phase solvent, the smaller inj volume will yield more narrow area of activity.

Here is an example of what I'm referring to. The is a C4 4.6 mm ID column with CH3CN + 0.1% HFBA in the mobile phase.

Gradient:
5-30% CH3CN over 5 min
30-50% CH3CN over 60 min

1.5 minute fractions collected

[pipettemonkey]

There is not an on-off difference between small molecules and proteins. The difference is gradual. The one adsorption/one desorption step for proteins is a drastic oversimplification, which can lead you astray, if you believe it to be literally true.

The real isue is the steepness of a plot of the natural logarithm of the retention factor versus the solvent composition. Small molecules have values around 10. Peptides have values around 20 to 40, and proteins have values around 80 or higher. 80 is 8 times steeper than 10, but it is not "on-off". A large injection volume for a peptide will also result in a larger elution volume. Not as bad as if this would have been for a small molecule, but you will still see measurable effects.

Is steepness of plot the polycratic regression coefficient, S?

Would peptides between 30-50 amino acids behave more closely to proteins, or short peptides?

Does injection volume affect peak shape in ion exchange as well?

[Uwe Neue]

The numbers that I referred to are based on the natural logarithm, so they are about 2.5 times larger than Snyder's S.

30-40 AAs give you a steepness value closer to proteins maybe about 60... but I am working from memory and need to look it up to be sure.

The injection volume will affect peak shape in ion-exchange even more than in RP, unless the charge of the protein is very high.