ask for advice on peptide mapping - exchange buffer, visual comparision, quality criterion (Jul/12/2011 )

Hi everyone!

I am encountering some problem with my peptide mapping for monoclonal antibody.pls help me if you have any idea,thank you!

My reduced and alkylated antibody appear precipitation when exchanged in 100mM NH4HCO3. This phenomenon don't occur at every time and some antibody may occur often. However it become clear after 16h trypsin digestion at 37 degree! And it seems OK in peptide map. But I still worry about the reproduciblity and looking for buffer to keep my antibody in solution. I tried adding 20mM salt in buffer, but it didn't work. I am wondering if you encountered similar problem, and how do you solve?

I have another question about result comparision. There are many peaks in the map, do you compare every peak with standard? I mean, there are still very tiny peaks, with small peak area percent, may have different behavior between sample and standard. And we have no MASS for detecting what in it now, but I should tell QC people which peaks we should integrate as a peak! And what level difference is thought as a change in sequence?0.1%? or 1%? or something else? There are no confirm
guidlines in pharmacopeia(USP. EP. JP). I have no idea. Could you give me some suggestions or just tell me how you do usually?

Looking forward your reply and discussion!

I think you have precipitation because your reduced and alkylated antibody is randomly refolding in buffer without urea or guanidine hydrochloride. Check how much urea or Gu-HCl your trypsin can withstand and exchange into buffer which contains one of those in amount within acceptable limit.

Do you have amino acid sequence of your antibody? If yes - make a theoretical digest (eg. using Expasy PeptideCutter) - that would give you the "correct" number of peaks. By checking AA composition you can tell how many of those would be visible on 280nm (while you see all of them on 210-200nm). I would assign those peaks starting from the largest (if you have only UV data) and say everything else is just a product of random and unspecific digestion.

You should distinguish between sequence change and sequence coverage (ie. how much of the sequence is actually detected) by the method. I mean - depending on the digestion products and chromatography condition you may loose some minor peptides which would lead to less than 100% sequence coverage - it does not however indicate actual change in the sequence.

Personally I would not state the change in the sequence without MS (or better - MS/MS), unless there's major change in HPLC-UV profile.

Thank you very much!

I added different amount GuHCl in buffer to check the resolution of the reduced and alkylated MAb these days, it seems work.

Yes, I have AA sequence of my antibody. And I maked the theoretical fragment by a sofware (GPMAW)(the result is same as Expasy). Peaks of my peptide map are much more than the theoretial ones. And the peaks are not small enough to be ignored. I think the unspecific digestion maybe the major reason, but I don't know how to identify those unspecific peaks with the existing condition. You just mentioned "By checking AA composition you can tell how many of those would be visible on 280nm (while you see all of them on 210-200nm). I would assign those peaks starting from the largest (if you have only UV data) and say everything else is just a product of random and unspecific digestion." Do you mean you only compare peaks that are visible at 280nm? if there are peaks from random and unspecific digestion, how you deal with these peaks when you perform method validation? Do you integrate them as peaks and caculate for area?

Thank you again!

K.B. on Tue Jul 12 18:15:50 2011 said:

hello sherry... as i can understand you are working on a biosimilar.. the answers to your questions depend on what is the objective of your method!!! are u using it as an identity method alone? or it is solving a characterization purpose. Non-specific cleavages in a tryptic map is common, more so in an antibody preparation. You may also try using other enzymes, most suitable would be Lys C as it can even overlap with a tryptic digest and you may get valuable information just by the two maps on LC ( as u mention MS is out of scope as of now. you may also try different reduced and alkylation conditions to compare the extent of non-specific cleavage of your antibody!!

Hope it helps..

Thanks a lot!
Yeah, we are working on a biosimilar, and peptide mapping is used in our lab as an identity method. I used to see both enzyme (trypsin and lys C) were applied in this method. lys C is purchasing now.I want to say, non-specific cleavage may exist in Lys C as it in trypsin, it is unavoidable and unpredictable. So it is a big difficulty for method validation. By the way, what is your strategy about this identity method? what parameters will you inspect? Retention time may be not enough.
I have confirmed that the redueced and alkylation condtion is sufficient and next step, I will try different digestion condition for compare the extent of non-specific digestion. Thanks for your suggestion!

Prep! on Mon Jul 18 04:53:52 2011 said:

yeah retention times are not enough... as u must be aware, the two basic parameters that shud be evaluated would be the specificity and the precision of the method. The acceptance criteria should focus more so on the relative persentages of major peaks (i would define major peaks as peaks susceptible to changes due to product related impurities generation - which forms a part of the method development) as the peak heights and areas may differ due to concentration effects. As long as the non-specific cleavages are happening in your reference standard as well, there should not be much cause for worry as far as identity of the molecule is concerned. But any peptide mapping method should be well characterized in the long run.

Hope that answers some of your concerns.. and i should share.. peptide mapping is the most difficult method to validate.. atleast till now what i have experienced!!! how much ever u do.. u can feel we can do more!!!

Million thanks! I will try my best, thanks for your kindly sharing!

Prep! on Mon Jul 18 07:26:40 2011 said:

yeah retention times are not enough... as u must be aware, the two basic parameters that shud be evaluated would be the specificity and the precision of the method. The acceptance criteria should focus more so on the relative persentages of major peaks (i would define major peaks as peaks susceptible to changes due to product related impurities generation - which forms a part of the method development) as the peak heights and areas may differ due to concentration effects. As long as the non-specific cleavages are happening in your reference standard as well, there should not be much cause for worry as far as identity of the molecule is concerned. But any peptide mapping method should be well characterized in the long run.

Hope that answers some of your concerns.. and i should share.. peptide mapping is the most difficult method to validate.. atleast till now what i have experienced!!! how much ever u do.. u can feel we can do more!!!

Best luck!!