Structural and isotope-heavy modifications occurring on the same residue

Structural and isotope-heavy modifications occurring on the same residue lpino  2021-06-30

Hi team! I've been banging my head against this one for a little while, so I thought I'd reach out to see if you can save me from any more headache!

Context: histone PTM analysis with D3-acetyl chemical derivatization. This means peptides will have biological (structural) modifications (including "normal" acetyl), and then any unmodified lysines and peptide N-termini should be D3-acetylated.

I did a paired DDA+DIA for each sample. Searched the DDA with Byonic (PD), built a spectral library with Bibliospec (Skyline), and added the detected peptides to the Skyline doc for quant by DIA XIC (Library Explorer > Associate proteins > Add).

Okay here's where I'm getting turned around. I am getting the craziest permutations of modification + D3 in the target list, and can't for the life of me interpret them! Some of these peptides have the same residue doubly-modified, e.g. a lysine with both structural "Acetyl" and then also isotope heavy "Acetyl:2H(3)". They all have a library spectra associated with them, so is this perhaps Skyline forcing all lysines to be "Acetyl:2H(3)" in addition to any other structural modification they might have?

I have no idea what I did wrong, but this is admittedly the deepest foray I've ever made into PTM analysis and I'm probably hacking some features together to get this to work. Is there some other way I should set up my Peptide Settings > Modifications? Any ideas or help would be greatly appreciated! Minimal Skyline doc example attached.

Thank you!

Brendan MacLean responded:  2021-06-30

Hi Lindsay,
I don't totally follow, but are you actually expecting light and heavy paired precursors in this case? It sounds more like you are saying that you want structural modifications only, but that some of them will have D3 and some will be "normal". If this is the case, then you would define multiple variable modifications with and without D3, and not use isotope modifications at all. Isotope-only modifications imply that you are expecting two structurally matching molecules that differ only by stable isotope labeling, which doesn't really sound like what you are describing.

If you are just trying to achieve a single structural molecule that happens to contain isotope labeling, and don't necessary expect its equivalent unlabeled form, or have any desire to measure it, then you can do that with structural modifications.

If I am wrong about this, can you point to exactly which peptides are problematic in your document and why. That may help me understand better.

Hope we can help achieve what you are trying for without requiring manual definition of the modifications on your peptides.


lpino responded:  2021-07-01

Right! I think you've got it! I wouldn't expect paired L/H precursors in the sense of normal SIL paired peptides, but yeah some will be D3 and some will be "normal". There may be situations where a peptide is biologically modified on some proteins and unmodified on other proteins in the sample, which would appear like a pair. For example, in our workflow we have some protein that has 2 PTM proteoforms, we prepare the samples (D3-acetylate any free lysines and the peptide N-terminus):

Proteoform 1 (doubly acetylated): PEKacTIDKacER is chemically derivatized by sample processing to become P(D3-ac)EKacTIDKacER
Proteoform 2 (unmodified entirely): PEKTIDKER is chemically derivatized by sample processing to become P(D3-ac)EK(D3-ac)TIDK(D3-ac)ER

In the document, the problematic peptides are all the ones with a [*] because it looks like in the Modification menu (right click on the peptide > Modify) these residues are listed as both structural (ac) and isotopic (D3-ac), which I don't think is even possible, e.g. to have a single lysine residue be doubly-acetylated?

It sounds like I'll try removing the D3-acetyl modification from the isotope modifications (I believe this was automatically added from the spectral library when I built it), and make a new D3-acetyl structural modification and see if that fixes it!

lpino responded:  2021-07-01

Posted too soon, should've poked around a bit more.

So I think you've got it, Brendan. Skyline is "splitting" the D3-acetyl modification into a structural acetyl and an isotopic D3. I'm going to try making a structural D3-acetyl mod, rebuild the spectral library, and see if that takes care of it!

lpino responded:  2021-10-10

Hello, sorry, I'm back!

I tried rebuilding the library after making those structural modifications (instead of the isotope modifications) and it looks like maybe part of my problem is that the mods are hardcoded (?) into the .msf/.pdResults files I'm using the build the library. After I rebuild with those structural mods added, when I go to "Explore Library", I get the Add Modifications popup which informs me that "This library appears to contain the modifications listed below. please sleet the ones you wish to use with the library:" and it lists those isotope modifications I tried to get rid of like Acetyl:2H(3) (H) and Dimethyl:2H(6) (K).

So I'm stuck again! Is there some way to force Skyline to accept/recognize [+45.029] as a heavy acetyl?

Nick Shulman responded:  2021-10-10
The only information that Skyline reads from a spectral library is the masses of the modifications on the amino acids.
Then, Skyline looks at the modifications that are already in the document. These are the modifications that you see at:
Settings > Peptide Settings > Modifications
Skyline tries to figure out what combination of structural and isotope modifications might be applied to any particular amino acid in order to equal the mass that is in the spectral library.

If Skyline is choosing the wrong modifications, then I think what you need to do is go to:
Settings > Peptide Settings > Modifications
and push one of the "Edit List" buttons and remove the modifications that you no longer want Skyline to consider.

When you do "View > Spectral Libraries", Skyline looks through all of the modifications that are already in the document. If Skyline cannot find any modifications in the document whose masses would explain the masses found on the amino acids in the spectral library, then Skyline looks through all of the unimod modifications that Skyline knows about, and suggests one of those. If you do not like the unimod modification that Skyline is suggesting, and would prefer a different modification be used, then you should go to "Settings > Peptide Settings > Modifications" and make sure the modifications that you want are already added to the document.

When you accept the suggestions that Skyline offers about which modifications to use, those modifications get added to the lists at "Settings > Peptide Settings > Modifications". If you change your mind and decide that you did not want to use those suggestions, you can get back to the way things were before by using the "Edit List" buttons at "Settings > Peptide Settings > Modifications" to remove the modifications you do not want.

If you have more questions you should probably send us your file again.
-- Nick