arginine N-methyltransferase definitions

[gocentral]

Gillian in FlyBase asked me about the following terms (needed to curate PMID:14705965):

arginine N-methyltransferase activity ; GO:0016273 --%protein-arginine N-methyltransferase activity ; GO:0016274 --%protein-arginine N5-methyltransferase activity ; GO:0019702

term: arginine N-methyltransferase activity goid: GO:0016273 definition: Catalysis of the transfer of a methyl group from S-adenosyl-L-methionine to the epsilon-amino group of an arginine residue. definition_reference: GO:mah

term: protein-arginine N-methyltransferase activity goid: GO:0016274 definition: Catalysis of the transfer of a methyl group from S-adenosyl-L-methionine to the epsilon-amino group of an arginine residue in a protein substrate. definition_reference: PMID:12351636

term: protein-arginine N5-methyltransferase activity goid: GO:0019702 definition: Catalysis of the transfer of a methyl group from S-adenosyl-L-methionine to the delta-nitrogen atom of peptidyl-arginine residues. definition_reference: PMID:9873020

The definition of GO:0016273 states that the transfer is to an epsilon group. However in GO:0019702 the transfer is to a delta residue. This seems like a TPV.

Also, in PMID:14705965,. the authors characterise Drosophila protein arginine methyltransferases which form N(omega),N'(omega)-dimethyl-L-arginine. Therefore GO:0018216 is ok for a process term, but the current function terms describe addition of the methyl group to an epsilon nitrogen, which I don't think is correct.

Should the definitions of GO:0016273 and GO:0016274 be changed to remove reference to an epsilon group? Or am I misunderstanding?

I think the TPV problem also applies to process:

peptidyl-arginine methylation ; GO:0018216 --%peptidyl-arginine delta-N-methylation ; GO:0019701 --%peptidyl-arginine 5-methylation ; GO:0018181

The definition of GO:0018216 lists only 3 (omega) products, not including the products formed during the process of GO:0019701 and GO:0018181. I think the definition of GO:0018216 should be broadened.

thanks, Becky

Reported by: beckyfoulger

Original Ticket: "geneontology/ontology-requests/1978":https://sourceforge.net/p/geneontology/ontology-requests/1978

[gocentral]

Logged In: YES user_id=436423

I've asked John Garavelli to look at this item -- he knows all about modifications, so he should be able to enlighten us.

m

Original comment by: mah11

[gocentral]

Logged In: YES user_id=436423

OK, John G has spoken ... things do need some work.

Part of the problem is that different systems of numbering (lettering?) the atoms in arginine have been used. The "internal" nitrogen has been called both delta and epsilon (the former because it's adjacent to the alpha- carbon, the latter because it is in the "epsilon" position...). So the definitions for GO:0016274 and GO:0019702 might just be using different systems, without actually saying so. Furthermore, this means that "delta methylation" is ambiguous, so we should try to avoid using that phrase in primary term names. Strictly speaking, "delta-N-methylation" (as in GO: 0019701) isn't ambiguous, but it could be confusing nevertheless, so we may also want to change it to "N6-methylation".

According to John, there are four arginine methyltransferase activities known:

- transfer of two methyl groups to one of the terminal (omega) N atoms; known as omega-omega or asymmetric; usually happens in the nucleus

- transfer of one methyl group to each omega N atom; known as omega- omega' or symmetric; usually extracellular; this presumably is what that Drosophila protein does

- transfer of a methyl group to the atom known as delta-C or C5; most are bacterial

- transfer of a methyl group to the atom known as delta-N, epsilon-N or N6; most are bacterial

GO has process terms for all three of the N-methylations (and it'll be easy enough to add C-methylation), but the function terms are a bit messier. There are terms for three specific activities (see below), but their definitions clash with those of the parent and grandparent terms.

As you noticed, there's no generic term for transfer to omega-Ns; we also have to add terms for transfer to N6 and C5 (the N6 ones could use essentially the same defs that are now attached to GO:0016273 and GO: 0016274).

We probably ought to broaden the definitions of GO:0016273 and GO: 0016274, because I suspect the terms have been used according to their names rather than their current definitions.

id: GO:0016273 name: arginine N-methyltransferase activity existing def: "Catalysis of the transfer of a methyl group from S-adenosyl- L-methionine to the epsilon-amino group of an arginine residue." [GO:mah] new def: "Catalysis of the transfer of a methyl group from S-adenosyl-L- methionine to an amino group of an arginine residue." [GO:mah]

id: GO:0016274 name: protein-arginine N-methyltransferase activity existing def: "Catalysis of the transfer of a methyl group from S-adenosyl- L-methionine to the epsilon-amino group of an arginine residue in a protein substrate." [PMID:12351636] new def: "Catalysis of the transfer of a methyl group from S-adenosyl-L- methionine to an amino group of an arginine residue in a protein substrate." [PMID:12351636, GO:mah]

I also checked the Enzyme Nomenclature entries -- they're all very substrate-specific, and have corresponding GO function terms:

%histone-arginine N-methyltransferase activity ; GO:0008469 ; EC: 2.1.1.125 %[cytochrome c]-arginine N-methyltransferase activity ; GO:0016275 ; EC:2.1.1.124 %[myelin basic protein]-arginine N-methyltransferase activity ; GO: 0016277 ; EC:2.1.1.126

For the process terms, it looks like the definition of GO:0018216 should be broadened to include modification of any residue (e.g. "The process of methylation of any residue of peptidyl-arginine.") Otherwise, the process terms aren't too much of a problem. If we assume that "peptidyl-arginine 5-methylation" (GO:0018181) means methylation of C5, the children of GO:0018216 are mostly OK, tho we should spell out "C5" in the name for GO:0018181 and correct "N5" to "N6" in the def of GO:0019701. There's also the "delta-N" business mentioned above. We could add a term for protein C-methylation.

Original comment by: mah11

[gocentral]

Logged In: YES user_id=482748

Hi Midori, John. thanks for the explanations!

Just one thing to clear up my confusion regarding N5 vs N6.

I'm not sure that N5 should be changed to N6 in the def of GO:0019701. The structure in RESID labels the product as peptidyl-N5-methyl-L-arginine. They list delta-N-methylarginine as a synonym. And the structure shows that the internal N is methylated.

So I'm a little confused about which Nitrogen is N5 and which is N6.

thanks Becky

Original comment by: beckyfoulger

[gocentral]

Logged In: YES user_id=436423

John - is that N5 an error?

Arginine has C at position 5 and N at position 6 (numbering with the COOH carbon as 1) ... so I would've thought there's no such thing as "N5" ...

m

Original comment by: mah11

[gocentral]

Logged In: YES user_id=436423

comment from John G:

The definition for GO:0019701, with PMID:9792625, definitely refers to the same modification as GO:0019702, with PMID:9873020, and they should probably be merged if they are both process terms. The "N5" designation used in the definition of GO:0019701 should be retained to avoid the delta/epsilon ambiguity. The definition for GO:0016274, with PMID:12351636 which refers specifically to the modification defined in GO:0019919, is evidently wrong. The activity in PMID:12351636 is quite specifically a histone, so the term can be retained, made a child of GO:0019919, and the protein substrate specified as histone (or even histone H4) in the corrected definition.

Original comment by: mah11

[gocentral]

Logged In: YES user_id=436423

response to previous comment:

> The definition for GO:0019701, with PMID:9792625, definitely > refers to the same modification as GO:0019702, with > PMID:9873020, and they should probably be merged if they > are both process terms.

GO:0019701 is a process term; GO:0019702 is a function term. So we can and should keep both.

> The "N5" designation used in the definition of GO:0019701 > should be retained to avoid the delta/epsilon ambiguity.

I heartily approve of avoiding the delta/epsilon ambiguity. But now I've become confused as to whether that nitrogen is N5 or N6. Did I just start counting at the wrong atom? (If so, we'll use N5 throughout, and it will all be fine.)

> The definition for GO:0016274, with PMID:12351636 which > refers specifically to the modification defined in GO:0019919, > is evidently wrong. The activity in PMID:12351636 is quite > specifically a histone, so the term can be retained, made a > child of GO:0019919, and the protein substrate specified as > histone (or even histone H4) in the corrected definition.

Is the rewrite I put in the SF item OK? (assuming that GO:0016274 is fairly generic, and we add a child term for the activity connected with the GO:0019919 modification)

Original comment by: mah11

[gocentral]

Logged In: YES user_id=482748

Hi Midori,

The RESID nomenclature seems to call both the nitrogen and the carbon as position 5, so i will go with whatever the experts say on that one.

I like all your definition suggestions Midori, including broadening the def for GO:0016274. Becky

Original comment by: beckyfoulger

[gocentral]

Logged In: YES user_id=436423

> The RESID nomenclature seems to call both the > nitrogen and the carbon as position 5

ouch! but I'll go with what the experts say too

m

Original comment by: mah11

[gocentral]

Logged In: YES user_id=436423

One more note from John (result of an impromptu conversation on Friday):

We should end up with three activity terms in molecular function for the omega-N methylations:

1) addition of a methyl group to either of the (unmethylated) omega-Ns

2) addition of a second methyl group to the same omega-N as in the first reaction, resulting in 'asymmetric' methylation

3) addition of a second methyl group to the other omega-N, resulting in 'symmetric' methylation

Known proteins possess activity (1) and either (2) or (3), never all three.

Becky - do you want to implement all this, or shall I?

m

Original comment by: mah11

[gocentral]

Logged In: YES user_id=482748

I'm happy to make these edits. Based on all comments in this SF entry, this is a summary of the changes that I think need doing:

1/ DEFINITION TWEAKS for the following terms:

term: arginine N-methyltransferase activity goid: GO:0016273 definition: Catalysis of the transfer of a methyl group from S-adenosyl-L-methionine to the epsilon-amino group of an arginine residue. definition_reference: GO:mah > definition: Catalysis of the transfer of a methyl group from S-adenosyl-L-methionine to an amino group of an arginine residue. definition_reference: GO:mah # term: protein-arginine N-methyltransferase activity goid: GO:0016274 definition: Catalysis of the transfer of a methyl group from S-adenosyl-L-methionine to the epsilon-amino group of an arginine residue in a protein substrate. definition_reference: PMID:12351636 > Catalysis of the transfer of a methyl group from S-adenosyl-L-methionine to an amino group of an arginine residue in a protein substrate. definition_reference: PMID:12351636 definition_reference: GO:mah # term: peptidyl-arginine methylation goid: GO:0018216 definition: The process of methylation of peptidyl-arginine to form the following derivatives: N(omega)-monomethyl-L-arginine, N(omega),N'(omega)-dimethyl-L-arginine, N(omega),N(omega)-dimethyl-L-arginine. definition_reference: RESID:AA0067 definition_reference: RESID:AA0068 definition_reference: RESID:AA0069 > Addition of a methyl group to any residue of peptidyl-arginine. definition_reference: GO:mah # term: peptidyl-arginine delta-N-methylation goid: GO:0019701 definition: The methylation of peptidyl-arginine to form peptidyl-N5-methyl-L-arginine. definition_reference: RESID:AA0305 definition_reference: PMID:9792625 > The addition of a methyl group on to the internal nitrogen atom (also called delta nitrogen, epsilon nitrogen or N5) of peptidyl-arginine to form peptidyl-N5-methyl-L-arginine. #

2/

TERM NAME CHANGE, and DEFINITION EDIT:

term: peptidyl-arginine 5-methylation goid: GO:0018181 definition: The methylation of peptidyl-arginine to form peptidyl-5-methyl-L-arginine. definition_reference: http://www.uni-marburg.de/mpi/thauer/thauer\_res.html definition_reference: RESID:AA0272 > term: peptidyl-arginine C5 methylation synonym: peptidyl-arginine 5-methylation definition: The methylation of peptidyl-arginine on a the carbon 5 (C5) residue to form peptidyl-5-methyl-L-arginine. definition_reference: http://www.uni-marburg.de/mpi/thauer/thauer\_res.html definition_reference: RESID:AA0272 definition_reference: FB:bf #

NEW TERMS TO ADD:

N-methyltransferase activity ; GO:0008170 ; EC:2.1.1.- --%arginine N-methyltransferase activity ; GO:0016273 ; ----%protein-arginine N-methyltransferase activity ; GO:0016274 ------%protein-arginine omega-N monomethyltransferase activity ; GO:NEW ------%protein-arginine omega-N asymmetric methyltransferase activity ; GO:NEW ------%protein-arginine omega-N symmetric methyltransferase activity ; GO:NEW

protein-arginine omega-N methyltransferase activity ; GO:NEW1 Catalysis of the addition of a methyl group to either of the unmethylated terminal nitrogen atoms (also called omega nitrogen) in peptidyl-arginine to form omega-N-G-monomethylated arginine residues.

protein-arginine omega-N asymmetric methyltransferase activity ; GO:NEW2 Catalysis of the addition of a second methyl group to methylated peptidyl-arginine. Methylation is on the same terminal nitrogen (omega nitrogen) residue that was previously methylated, resulting in asymmetrical peptidyl-N(omega),N(omega)-dimethylated arginine residues. comment: Type I protein arginine N-methyltransferase enzymes possess activity GO:NEW1 and GO:NEW2.

protein-arginine omega-N symmetric methyltransferase activity ; GO:NEW3 Catalysis of the addition of a second methyl group to methylated peptidyl-arginine. Methylation is on the terminal nitrogen (omega nitrogen) residue that is not already methylated, resulting in symmetrical peptidyl-N(omega),N'(omega)-dimethyled arginine residues. comment: Type II protein arginine N-methyltransferase enzymes possess activity GO:NEW1 and GO:NEW3.

C-methyltransferase activity ; GO:0008169 ; EC:2.1.1.- --%peptidyl-arginine C-methyltransferase activity ; GO:NEW

peptidyl-arginine C-methyltransferase activity ; GO:NEW Catalysis of the transfer of a methyl group to the carbon atom of an arginine residue in a protein.

peptidyl-arginine methylation ; GO:0018216 --%peptidyl-arginine N-methylation ; GO:NEW ----%peptidyl-arginine delta-N-methylation ; GO:0019701 ----%peptidyl-arginine omega-N-methylation ; GO:NEW ------%peptidyl-arginine methylation, to symmetrical-dimethyl arginine ; GO:0019918 ------%peptidyl-arginine methylation, to unsymmetrical-dimethyl arginine ; GO:0019919

--%peptidyl-arginine C-methylation ; GO:NEW ----%peptidyl-arginine 5-methylation ; GO:0018181

peptidyl-arginine N-methylation ; GO:NEW The addition of a methyl group onto a nitrogen atom of an arginine residue in a protein.

peptidyl-arginine omega-N-methylation ; GO:NEW The addition of a methyl group onto a terminal nitrogen (omega nitrogen) atom of an arginine residue in a protein.

peptidyl-arginine C-methylation ; GO:NEW The addition of a methyl group onto a carbon atom of an arginine residue in a protein.

Q1: What was decided about GO:0019701 (peptidyl-arginine delta-N-methylation). Should this be changed to peptidyl-arginine N5-methylation, with the existing text string being made a synonym?

Q2: Regarding the function terms for omegaN methyltransferases:

Catalysis of monomethylation (where the substrate is unmethylated) = 1 activity. Does catalysis of (asymmetric/symmetric) dimethylation refer to just the second methylation activity (ie methylation to a monomethylated substrate- as in the suggested definitions for the above new term) or does one function term cover both the first and the second methylation steps? From Midori's last comment I assume the former, but just wanted to check.

let me know if you think we need any more edits/new terms/tweaks etc that aren't listed above, and I'll do this all at the end of the week.

thanks, Becky

Original comment by: beckyfoulger

[gocentral]

• assigned_to: nobody --> beckyfoulger

Original comment by: beckyfoulger

[gocentral]

Logged In: YES user_id=436423

Thanks, that will be fantastic :)

Replies to queries ...

> Q1: What was decided about GO:0019701 > (peptidyl-arginine delta-N-methylation). Should this > be changed to peptidyl-arginine N5-methylation, with > the existing text string being made a synonym?

I think so.

> Q2: Regarding the function terms for omegaN > methyltransferases: [snip]

There should be one function term per step -- the proposed terms and definitions look just right.

... and one more thing: I think we should change 'unsymmetrical' to 'asymmetrical' in GO:0019919 -- better English and consistent with the other terms.

Original comment by: mah11

[gocentral]

Logged In: YES user_id=482748

Did all the edits described below. Also changed the text string of GO:0019701 to peptidyl-arginine N5-methylation, and changed 'unsymmetrical' to 'asymmetrical' in GO:0019919.

New terms added are: GO:0035241 protein-arginine omega-N monomethyltransferase activity GO:0035242 protein-arginine omega-N asymmetric methyltransferase activity GO:0035243 protein-arginine omega-N symmetric methyltransferase activity GO:0035244 peptidyl-arginine C-methyltransferase activity GO:0035245 peptidyl-arginine C-methylation GO:0035246 peptidyl-arginine N-methylation GO:0035247 peptidyl-arginine omega-N-methylation

Original comment by: beckyfoulger

[gocentral]

• status: open --> closed-fixed

Original comment by: beckyfoulger