Protein analysis by HPLC

[Perachiotti Anna]

During HPLC analysis of a protein (molecular weight about 30.000) with active center being serine, we had problem as the area of our peak tent to significantly increase with time (over 10% in few hours). We are absolutely sure of our HPLC performances and the problem is not due to the instrument.
We tried to dissolve standard and samples with differen pH and buffers (EDTA 0.02M, KH2PO4 0.02M, NaCitrate 1M, NH4Acetate 1M ) but the problem remains. Furthermore the area of the sample is always much higher than the standard area

[HW Mueller]

What makes you so sure of your LC performance? Looks like you are "covering" some active sites, early on, which would mean you loose some protein early. It could also be carryover or contamination.

[Perachiotti Anna]

Hallo,
thank you for your suggestions.
We can exclude carry over because injection of blank after samples are clean.
WE can exclude contamination because placebo solutions (tablets without Active principle, which is our protein) prepared in the same way of the samples are clean. Moreover blank prepared in the same way and with the same solutions of samples are clear.
I do not know about covering column active sites or loosing protein early. How do you suggest to act in order to understand if this is the problem ?
Moreover, we had the feeling that our protein was not stable in solution and its active site could change with time but we try different ways to stabilize it but we did came up always with increasing areas.
(The protein is Semi-alkaline proteinase SAP)
We are very grateful if you could help us in some way.

[Chris Pohl]

Perachiotti Anna,

You don't mentioned the nature of your stationary phase or your analytical buffer system. The problem you describe could potentially be related to long equilibration times required for the stationary phase pH to achieve equilibrium with the mobile phase pH. Depending upon your buffer system and the nature of your stationary phase there can be significant differences between the mobile phase pH and the stationary phase pH.

[HW Mueller]

I assumed RP or SEC. If your blank doesn´t match your sample closely you may be in trouble, regarding seeing carryover.
With active sites in the column I meant those which might clamp on to the protein (early refers to the first few injections). It seems that chaotropics, etc., can sometimes eliminate such problems. or else: some people just "load" those sites by injecting the protein until it doesn´t change concentration anymore. I don´t like the latter, it could be non-robust (equilibrium shifts). Now, protein loading can probably not be used with gradients (in RP).
On Chris´s buffer non-equilibrium: maybe at the low buffer concentrations? At 1M I would not expect it in RP or SEC, ..possible in ion exchange?

[kiknos]

Hallo,

We can exclude carry over because injection of blank after samples are clean.
.

i wonder when you did such trial, immediately after the original several injections, or some time("with hours"...) later when the apparent bigger peak area has been observed?

[Perachiotti Anna]

Our Stationary phase is RP 18 300A
Our Mobile phase was K2HPO4 0.02M /CH3CN = 95/5 pH 6.8
WE will try today a different mobile phase KCL 0.1M / CH3CN = 95/5 (it shoul help to inactivate active column sites)

Concerning loading of column: we did try to load higher quantities but may be the colum was not loaded enough (our sample is 4000 y/ml and we loaded three injections 4 times comcentrated. Is this enough ?)

Concerning blank we made the injections after the bigger peaks were observed.

Have you any idea why our sample, when prepared in the same why of the standard, at the same concentration of the standard , show areas nearly double ?

thanks a lot

[HW Mueller]

It is very simple to know whether you have loaded (covered, equilibrated) enough: Your peaks no longer increase on subsequent injections.
I don´t understand your last question, you know how much is in your sample (why do you analyze?)? Or are you talking about spiking of the sample? Could it be that your sample is in a medium that "helps" it to get through the column?
Do I understand correctly that you are running this isocratically? Your protein does not come out at tm? Is the peak shape of "sample" and St the same?

[Perachiotti Anna]

We run an isocratic and the peak shape of Std and sample is the same.
We know the sample concentration because we are working on pharmaceutical tablets manufactured by our prodaction site and we know by production weight how much Active principle we have in each tablets. Moreover we have an enzimatic method (very long and time consuming) which has been validated and give us the precise amount of Active Principle in the sample. We would like to set an HPLC method as it should be faster for routine analysis. We analyse aur sample even if we know the content because regulatory and Ministry of Health needs to know the exact percentage (95%-105%) of Active Ingredient.

[HW Mueller]

It looks very much as mentioned before: your standard is most likely in the wrong solvent and partially sticks somewhere.
It is still interesting to know whether your protein appears at the tm (dead time, etc.).
What happens to the peak area when you mix equal volumes of St and sample and inject the mixture?

[Uwe Neue]

Hmm... you are doing an isocratic run with a protein on a 300 A C18 packing. This is very unusual, and the mere fact that you get a peak makes me suspect that the peak is unretained. What is the retention time at which flow rate, and what is the column dimension?

[Perachiotti Anna]

The retention time is 8 minutes and the flow 1 ml / min.

concerning sample areas higher than std areas we noticed that when adding to our standard the excipients of the coated tablets the std area increase and gets similar to the sample area. So the increasing areas of the samples are probably due to some action of placebo excipients on our protein. We think that is a problem without solution as it would be a nonsense addind to our standard the placebo excipients each time we need to analyse the tablets.

[Einar Ponten]

You may need to do calibration by standard addition instead.

Maybe there is a possibility to decrease the injection volume and dissolve/ dilute both sample and standard in mobile phase. It seems that it may improve the situation. Tablets are likely to contain compounds that helps to solvate the protein and keep it from surfaces of containers, vials, tubing, syringes etc.

[HW Mueller]

Assuming a 4.6x250mm column (or shorter) it looks like you really have retention. That is surprising. How do you know that what you see is the main "active" and not a small peptide (or???) in the "active", the main protein being hung up on the column?

[Uwe Neue]

And what are the column dimensions?