Hi Josephine!

For starters, predicted glycosylation sites are just that: predicted. Usually glycosylation, especially of not 'naturally' glycosylated proteins, is not 100% efficient; I'd say very far from that.  O-linked glycans are quite smaller than N-linked glycans (I assume we talk eukaryotic cells, right?), usually 1 up to 4 sugars of 164 or 204 of mw.

Regarding your bands, is it possible that your lower Mw band on secreted protein is due to the cleaved export signal? it would be my choice, run a predictor to see your signal peptide and subtract its Mw to the one of the full (unglycosylated) protein.

The upper band (secreted) could be your protein with glycan attached (keep in mind that glycans are not peptidic, so their size may not reflect on mobility like proteins). Depending on how deep you want to look into this you have 2 choices (at least):

a) Do a western with anti-glycan antibodies (those are commercially available), and you should see a signal only on your higher band of those 3.

b) Treat your protein sample with a glycan cleaving enzyme, they are commercially available (I think from Sigma, NEB, etc) and you should "restore" the Mw of the secreted protein to the lower size (signal cleaved, but no glycan).

There are also less expensive methods to check for presence of sugars, but would require more "setup" time.

Hope it helps and good luck!

Nico.

PD: there's some folks at ICL that might have good insights on this :)

Hi Josephine!

For starters, predicted glycosylation sites are just that: predicted. Usually glycosylation, especially of not 'naturally' glycosylated proteins, is not 100% efficient; I'd say very far from that.  O-linked glycans are quite smaller than N-linked glycans (I assume we talk eukaryotic cells, right?), usually 1 up to 4 sugars of 164 or 204 of mw.

Regarding your bands, is it possible that your lower Mw band on secreted protein is due to the cleaved export signal? it would be my choice, run a predictor to see your signal peptide and subtract its Mw to the one of the full (unglycosylated) protein.

The upper band (secreted) could be your protein with glycan attached (keep in mind that glycans are not peptidic, so their size may not reflect on mobility like proteins). Depending on how deep you want to look into this you have 2 choices (at least):

a) Do a western with anti-glycan antibodies (those are commercially available), and you should see a signal only on your higher band of those 3.

b) Treat your protein sample with a glycan cleaving enzyme, they are commercially available (I think from Sigma, NEB, etc) and you should "restore" the Mw of the secreted protein to the lower size (signal cleaved, but no glycan).

There are also less expensive methods to check for presence of sugars, but would require more "setup" time.

Hope it helps and good luck!

Nico.

PD: there's some folks at ICL that might have good insights on this :)

Hi!

Thank you for your swift response and for confirming some of the things we were suspecting!

We already ordered a deglycosylation kit (with 4 enzymes) last week so hopefully this will answer our question.

Regarding the lower molecular weight band in CM, this should only make a difference of 4 kDa, so it does not seem to be the cleavage of the signal peptide during secretion that is causing this. 

What we have been thinking to do is to basically run the samples on a gel, cut out the two separate bands and send them for 'protein sequencing', which I have heard is possible. Would this be done in a mass spectrometry facility? We're trying to find such a facility in London at the moment.

Best,

Josephine

Is it possible that your CL band is partially glycosylated, while the CM line has unglycosylated and fully glucosylated versions?

Since this is a cartoon, it is difficult to help you with analysis as it is hard to determine how big are the shifts etc. A picture of a real SDS PAGE gel wold be better.

Hi guys,

I think you are on the right track there! Mass spectrometry will tell you the protein sequence and it can even tell you if the protein is glycosylated, we used to do it all the time here. It is strongly dependent on the expertise of the people doing it, so they know what to look for and how to spot it. I can look around a little bit in the London area if you don't know anyone; I used to work with people who collaborated with scientists in London.

I think Konrad may also have a point, since the two glycosylation pathways are quite independent you may have just one of them in certain conditions. If your proteins are his-tagged, I would purify from both conditions (CL & CM), run them on SDS-PAGE, cut the bands and proceed to Mass spectrometry identification, you don't need much amount for it, usually 1 ug should be enough. The protocol can be obtained from Shevchenko et al (PNAS 1996) or I have a chapter on protein O-glycosylation, following roughly the same protocol.

Hope it works, keep me posted.

Nico.

Hi Nicolás and Konrad,

Thank you very much for both of your responses. I am sorry I am sending such a delayed response, I forget that research gate doesn't send me any notifications to my email (which can be refreshing at times considering basically all other sites do..)

Konrad; is there a biological explanation to why glycosylation would be lower in the secreted sample? Recently, we have actually seen a stronger lower band (which could indicate degradation as well) and only a weaker upper band.

Nicolás, thank you for offering to contact people in London, I have contacted a mass spec facility at Imperial College and they have agreed to look into it.

We have been using our His-tag to purify on a Ni-NTA column, but it has not been pure enough in the first efforts (a great PhD student in the lab has tried), but we now have magnetic DynaBeads that we plan to use with a myc antibody. (The protein has both a myc and a His tag)

Thanks for your help, both of you!

Hi Josephine,

Sorry but I can't help you with that, glycosylations pathways are very complex and I am not an expert on that.

I think Nicolas is right, both pathways are different, each one can be partial or completely glacosylated. To make things even more complicated, you can have a combination of both, different form of glycosylation and partial clipping of your protein.

Are both tags on the same terminal?

Can you detect them both with Western Blot on all three bands?

Mass Spec should answer all questions.

Let us know what you find out.