PANTHER:PTN000109455

[ValWood]

from
#1873

regulation of transcription, DNA-templated
histone arginine methylation
histone-arginine N-methyltransferase activity

we have no evidence that histone is a physiological substrate ( I think we would know this in f.y. for this particular methyltransferase)

I will check up on this one....

[thomaspd]

@ValWood , do you think these PAINT annotations are OK for now, and we can close this issue?

[ValWood]

It might be OK. Close, and I'll follow up and open a new ticket if it isn't.
It seems odd. I think we would know if it was a histone modifier in fission yeast, but perhaps not...

Remember that often people use histones as substrates in "in vitro" experiments, but they are not physiological substrates. Happens a lot with kinase assays too. Is this annotation from such an experiment?

[ValWood]

This is what we have for rmt1 in PubMed:

https://www.ncbi.nlm.nih.gov/pubmed?cmd=Search&term=(SPAC890.07c+OR+rmt1+OR+prmt1)+AND+(pombe+OR+fission+yeast)

https://www.pombase.org/reference/PMID:15175657

I'm going to ask Francois (author) if he thinks histones are a likely substrate....

[ValWood]

From François:

"Although it is very likely that Rmt1 would modify histone tails, I doubt this was demonstrated by either mass spec or Western blotting using extracts of S. pombe. Not even sure that S. pombe Rmt1 was shown to methylate histones in vitro."

So basically, it is likely that this reaction would occur in vitro it has not been demonstrated, in vivo or even in vitro.

[thomaspd]

Thanks for this-- it's so nice to have an expert on speed-dial!

So for the PAINT annotations, do you think it's OK to leave histone as a predicted (in vivo) substrate, or do you feel that is likely enough to be incorrect, that we should look into removing it (or restricting it to a narrower clade)?

[ValWood]

Thanks for this-- it's so nice to have an expert on speed-dial!

I know, we have a few >600 community curators on speed dial now, It's really helpful. We are user-driven. It is the future!

So for the PAINT annotations

It depends on the context of the experiments - was histone only used as an acetylation assay? PomBase doesn't annotate in vivo substrates if we know they are not likely to be physiologically relevant.

But as far as I can see in most organisms this is a polyA binding methyltransferase.....

[thomaspd]

Thanks Val. @marcfeuermann can you please take a look at substrate specificity across this family and decide how to annotate the gains and losses?

[marcfeuermann]

Enzyme activities are usually demonstrated in vitro (most of kinase substrates have been shown like that). This of course questions the physiological relevancy of those experiments. Just looking at the PMRT1-branch which includes pombe rmt1, there is in vitro evidence for Human, Mouse, Rat, D. melanogaster, C. elegans, A. nidulans, P. falciparum and even S. cerevisiae (not all of these have the GO annotation). Human PRMT1 has been shown to be the main enzyme that methylates H4R3 in vivo (PMID11448779).
Moreover, there is specificity of the "PRMT branches" toward some histones (in vitro artifact ?)
Subcellular localization to nucleus of those proteins is also coherent with this (I agree it is not a proof and by far my weakest argument).
And finally, even the expert says "Although it IS VERY LIKELY that Rmt1 would modify histone tails, I doubt this was demonstrated by either mass spec or Western blotting using extracts of S. pombe. Not even sure that S. pombe Rmt1 was shown to methylate histones in vitro."
I understand your concern Val, but with all these data, I really would keep the inference that histones are substrates, even if it has not been demonstrated in vivo in pombe. This does not mean that these enzymes exclusively methylate histones.

[ValWood]

OK, I'm happy for the histone acetylation to stay. Although I interpreted Francois response to say
"he thought it could likely acetylate histone in vitro artefactually, but he hadn't seen it".
So, we can keep this but since not every histone modification is transcriptional regulation, was there strong evidence for this BP?

The in vitro paper you refer to above says "These findings suggest a role for arginine methylation of histones in the transcription process.

I'm not sure that the papers used for transcription really show a definite role in transcription
https://www.uniprot.org/citations/19472346
and one of the evidences is from mouse calm1 which is not a direct ortholog of pombe Rmt1

I would be more conservative about a transcription annotation without good evidence because mutations could affect transcription indirectly.

But we can keep this for now if you think it's OK.

Cheers

Val