Glycan analysis

[biochemist]

We are planning to begin glycan analysis of recombinant proteins (both for routine QC, and for simpler characterization tasks).

Right now we have Waters 2695 pumps with fluorescent and UV detectors. No centrifugal evaporators, or UPLCs. We have normal equipment like ovens, water baths, etc


Can anyone share their experience with the following:

1) What additional equipment we'll be needing

2) What type of labelling is best (2AA, 2AB, DMB etc) for neutral glycans; and for sialic acids

3) Column recommendations?

4) Kit recommendations?


The methods should be robust and reproducible enough for a QC lab to use... thanks!

[Alexandre]

I found that amino columns are better that hilic for non derivatised sugars. For identification we used UPLC and nanoLC/Orbitrap. You need the best possible separation. If you don’t want to spend too much, try at least sub-3um columns. QC should be possible to do for fixed markers the simple way - kit or no kit/deriv./HPLC/FLD.

[biochemist]

Thanks Alexandre for your reply.

If you don’t want to spend too much, try at least sub-3um columns.

Can I use sub-3um columns with my current detectors (2996 PDA, 2475 fluor)? I think they have the standard flow cells..

I found that amino columns are better that hilic for non derivatised sugars.

For non-derivatised sugars, do you mean with PAD detection? Or MS?


Is there an alternative for sample drying/ concentration steps for glycan analysis if I don't have a Speed Vac? Is it possible to just use a UF cartridge?

[SiPeel]

Hi biochemist,

First off I have to declare my interest, I work for a glycoprofiling company and I think you may do well to look at our website. It's www.ludger.com.

I think you'll struggle without the ability to dry down samples, specifically following release of the glycans in preparation for labelling. So I'd say get yourself a centrifugal evaporator if you can. Other than that you should be fine to run your analysis with your current HPLC equipment. In terms of labels, I would suggest 2AA over 2AB as it has a better response factor on HPLC. The only problems with that is that 2-aminobenzoic acid is a controlled substance so you can run into problems when buying it (I thik depending on quantites) and it's not as good for positive ion MS analysis if you wanted to do that too. We regularly perform DMB labelling for sialic acid analysis and find it a robust method, however stability of the dye is short so you have to analyse within a day or so of labelling.

In terms of columns and kits, we sell our own so I'm duty bound to say they are the best but I'm sure there are plenty other alternatives out there.

Good luck setting up your method

[lmh]

May I nominate SiPeel's reply as the most professionally-worded "with-interests" reply I have yet seen on this site? Congratulations!

[biochemist]

Yes, that's excellent advice. I'm going to be checking out the Ludger products.

Could you recommend a Ludger workflow I could follow for EPO? And mAbs?

And how can I estimate if my glycan analysis is accurate - i.e. whether I'm losing glycans (through enzymatic or non-enzymatic ways) during sample prep?

[cari]

Dear Biochemist,

If you have access to a CE instrument (which you should, regarding the area you work in), I'd also have a look at the glycan kits available from e.g. Beckman Coulter. The labelling is done with APTS, and the good news is that you do not need to remove all excess labelling material, since it does not interfere with the separation. The analysis is as fast as the UPLC method, but a lot cheaper per sample.
This methodology is being used in several QC labs from biopharma companies.

Good luck!
Cari

[biochemist]

Does the APTS increase the UV absorbance of the glycans - or are there other labels that do?

Unfortunately we only have a CE-UV and not a CE-LIF.

Also, another question - what is everyone's favourite PNGase F enzyme brand, and preferred method of clean-up?

[lmh]

APTS absorbs as well as fluorescing (however you spell that). Fluorescence is, of course, vastly vastly more sensitive.
Resolution is stunning (makes UPLC peaks look broad). CE is brilliant if you've got it.