issues with "copper ion sensor activity"

[gocentral]

Hi,

Basically, I’d like to make new terms that better represent the MF of the transcriptional regulator, csoR, that the term “copper ion sensor activity” was created to annotate. These new terms will be more consistent with the other copper binding txn factor activity terms I have created. Becky and I have discussed this a little bit offline, but I'll put all the background here.

Here is a question to consider in reading the information below.

Question: What should be the fate of the existing “copper ion sensor activity” term (GO:0097077)?

I am not keen on converting 97077 into the term, or even a parent of the term, that I think should actually be used for annotating csoR since the existing def doesn’t talk about the effect on DNA binding or txn at all, so it seems like a violation of our rules to make the term a lot more specific. So, it seems like 97077 should probably be obsoleted. Thoughts?

-Karen

Background:

There is a MF term "copper ion sensor activity" GO0097077: with this definition: Interacting selectively and non-covalently with and responding, e.g. by conformational change, to changes in the cellular level of copper(I) (Cu+).

The 97077 term predates this MF term I created as part of the transcription overhaul: Term: copper ion regulated core promoter proximal region sequence-specific DNA binding RNA polymerase II Def: Interacting selectively and non-covalently with copper(I) (Cu+) ion(s) and responding, e.g. by conformational change, to changes in the cellular level of copper(I) (Cu+), such a way that copper ion binding regulates the activity of the transcription factor, perhaps by regulating the ability to interact selectively and non-covalently with a sequence of DNA that is in cis with and relatively close to a core promoter for RNA polymerase II (RNAP II), or by regulating some other function of the transcription factor, in order to modulate transcription by RNAP II. GOID: GO: 0001202

When I created the "copper ion regulated ... txn factor activity" I gave it "copper ion sensor activity" (97077) as an is_a parent because based on the def of "copper ion sensor activity", the term I created (1202) was acting in that manner. However, this threw an error for QuickGO:

From Becky’s email on 29 Nov 2011: Tony noticed a slight problem with the following new GO term:

copper ion regulated core promoter proximal region sequence-specific DNA binding RNA polymerase II transcription factor activity ; GO:0001202

It's blocking a QuickGO update because of a loop (it's reaching 'copper ion binding ; GO:0005507' via both a HAS_PART, and an is_a relationship):

copper ion binding ; GO:0005507 --[isa]copper ion sensor activity ; GO:0097077 ----[isa]copper ion regulated core promoter proximal region sequence-specific DNA binding RNA polymerase II transcription factor activity ; GO:0001202

copper ion binding ; GO:0005507 --[HASPART]copper ion regulated core promoter proximal region sequence-specific DNA binding RNA polymerase II transcription factor activity ; GO:0001202

As a short term fix, Becky removed the is_a link between GO:0097077 and GO:0001202 as a quick fix (so QuickGO can update) since the HAS_PARTs are consistent with other TF terms.

Proposal:

Regarding the 97077 “copper ion sensor activity”, it was created from PMID: 19928961 and, in AmiGO, has only been used to annotate one gene product, csoR, from Mycobacterium tuberculosis and also a structure from the same species, likely corresponding to the same gene product since both annotations use the PMID: 19928961 that was used for the term definition. However, the term “copper ion sensor activity” with its current definition only captures half of the functional information in PMID: 19928961. The gene csoR is a transcriptional repressor that binds DNA in the apo-copper form, but which loses affinity for DNA when it binds copper.

In contrast, the CUP2 gene from S. cerevisiae requires binding to copper in order to bind DNA.

Both csoR and CUP2 have very similar functions, so I would like to represent the terms for the MF’s of these genes in a consistent way.

Since the way copper binding is affecting DNA binding is opposite for the two example cases, csoR and CUP2, I would like to indicate in the MF terms whether the copper binding is activating or repressing on DNA binding. Using the high level parent term that already exists (1199) as the base for my examples below, I would like to create terms like new1 and new2 below and corresponding specific children for copper in the bacterial-type and RNAP II branches:

GO:0001199 – metal ion regulated sequence-specific DNA binding RNA polymerase transcription factor activity

GO:new1 – metal ion activated sequence-specific DNA binding RNA polymerase transcription factor activity

GO:new1 – metal ion repressed sequence-specific DNA binding RNA polymerase transcription factor activity

Reported by: krchristie

Original Ticket: geneontology/ontology-requests/9305

[gocentral]

I agree you should obsolete GO0097077. The two new terms look good to me.

Original comment by: jl242

[gocentral]

• assigned_to: nobody --> kchris

Original comment by: jl242

[gocentral]

There's also this term that I created some time ago also from the same submitter: GO:0097063 cadmium ion sensor activity SF link: https://sourceforge.net/tracker/index.php?func=detail&aid=3313597&group\_id=36855&atid=440764 GO:0097063 has only been used to annotate CmtR from PMID: 19456862. The paper title reads "CmtR, a cadmium-sensing ArsR-SmtB repressor, cooperatively interacts with multiple operator sites to autorepress its transcription in Mycobacterium tuberculosis." ...so should this term also go? Or can it stay?

Thanks, Paola

Original comment by: paolaroncaglia

[gocentral]

Hi Paola,

From your description (not having looked at the paper yet), it sounds to me like the "cadmium ion sensor activity" term should treated the same way as the "copper ion sensor activity" term and I can make the appropriate copper specific and cadmium specific terms under the term "metal ion regulated sequence-specific DNA binding RNA polymerase transcription factor activity" (GO:0001199).

-Karen

Original comment by: krchristie

[gocentral]

Thanks Karen,

I'll let you double-check that with the paper then, whenever it's convenient for you. Let me know if there's anything I should take care of.

Cheers, Paola

Original comment by: paolaroncaglia

[gocentral]

I agree that GO:0097077 should be obsoleted and maybe the term name can become a synonym of the new terms. If the cadmium term is acting in the same way, then I think it should follow a parallel structure.

David

Original comment by: ukemi

[gocentral]

Thanks for commenting David.

The CmtR cadmium regulated txn factor is functioning essentially the same way as the copper regulated one I started this SF item about.

The CmtR paper also cited 2 interesting reviews, listed below. I skimmed the one about the MerR family. These are activators, but the prototype of the family, MerR binds to DNA in both the metal bound and apo states, but metal binding affects its ability to activate txn. Thus, will need to make sure my defs don't indicate that metal binding regulates DNA binding. Some do this, some regulate a different step.

Anyway, I think it's justified to obsolete both the copper and cadmium ion sensor activity terms. It's probably a good idea to use them as synonyms, possibly of the RELATED type, for the new terms where appropriate.

-Karen

Brown NL, Stoyanov JV, Kidd SP, Hobman JL. The MerR family of transcriptional regulators. FEMS Microbiol Rev. 2003 Jun;27(2-3):145-63. Review. PubMed PMID: 12829265.

Busenlehner LS, Pennella MA, Giedroc DP. The SmtB/ArsR family of metalloregulatory transcriptional repressors: Structural insights into prokaryotic metal resistance. FEMS Microbiol Rev. 2003 Jun;27(2-3):131-43. Review. PubMed PMID: 12829264.

Original comment by: krchristie

[gocentral]

Added new terms & made some minor alterations/corrections to def & term names of some old terms

terms with new parents GO:0001140 - bacterial-type RNA polymerase core promoter proximal region sequence-specific DNA binding transcription factor activity involved in positive regulation of transcription GO:0001141 - bacterial-type RNA polymerase core promoter proximal region sequence-specific DNA binding transcription factor activity involved in negative regulation of transcription

terms with minor modification to def GO:0001199 (name change also) - metal ion regulated sequence-specific DNA binding RNA polymerase transcription factor activity GO:0001200 - metal ion regulated sequence-specific DNA binding RNA polymerase II transcription factor activity GO:0001201 - metal ion regulated core promoter proximal region sequence-specific DNA binding RNA polymerase II transcription factor activity GO:0001202 - copper ion regulated core promoter proximal region sequence-specific DNA binding RNA polymerase II transcription factor activity GO:0001203 - zinc ion regulated core promoter proximal region sequence-specific DNA binding RNA polymerase II transcription factor activity GO:0001204 (name change also) - metal ion regulated sequence-specific DNA binding bacterial-type RNA polymerase transcription factor activity

new terms: GO:0001209 - metal ion regulated sequence-specific DNA binding transcription factor activity involved in positive regulation of transcription GO:0001210 - metal ion regulated sequence-specific DNA binding transcription factor activity involved in negative regulation of transcription GO:0001211 - copper ion regulated core promoter proximal region sequence-specific DNA binding RNA polymerase II transcription factor activity involved in positive regulation of transcription GO:0001212 - zinc ion regulated core promoter proximal region sequence-specific DNA binding RNA polymerase II transcription factor activity involved in positive regulation of transcription GO:0001213 - metal ion regulated core promoter proximal region sequence-specific DNA binding RNA polymerase II transcription factor activity involved in positive regulation of transcription GO:0001214 - metal ion regulated core promoter proximal region sequence-specific DNA binding RNA polymerase II transcription factor activity involved in negative regulation of transcription GO:0001215 - metal ion regulated sequence-specific DNA binding bacterial-type RNA polymerase transcription factor activity involved in positive regulation of transcription GO:0001216 - sequence-specific DNA binding bacterial-type RNA polymerase transcription factor activity involved in positive regulation of transcription GO:0001217 - sequence-specific DNA binding bacterial-type RNA polymerase transcription factor activity involved in negative regulation of transcription GO:0001218 - metal ion regulated sequence-specific DNA binding bacterial-type RNA polymerase transcription factor activity involved in negative regulation of transcription GO:0001219 - copper ion regulated sequence-specific DNA binding bacterial-type RNA polymerase transcription factor activity involved in negative regulation of transcription GO:0001220 - cadmium ion regulated sequence-specific DNA binding bacterial-type RNA polymerase transcription factor activity involved in negative regulation of transcription

Original comment by: krchristie

[gocentral]

• assigned_to: kchris --> ukemi

Original comment by: krchristie

[gocentral]

Could we make 'copper ion sensor activity' (GO0097077) a 'receptor activity' (it seems to fit the def) and make it has_part copper ionbinding?

David

Original comment by: ukemi

[gocentral]

There's several 'x sensor terms':

cadmium ion sensor activity ; GO:0097063 carbon monoxide sensor activity ; GO:0070027 copper ion sensor activity ; GO:0097077 nitric oxide sensor activity ; GO:0035991 oxygen sensor activity ; GO:0019826 sterol sensor activity ; GO:0032935

I agree that these should all be better with HAS_PART relationships to the binding term.

Is it just the copper and cadmium ones that refer to metal-ion binding transcription factors? Before we alter the sensor terms, are both the sensor term and the TF term needed for these?

copper ion sensor activity ; GO:0097077 copper ion regulated ... TF activity ; GO:0001202

cadmium ion sensor activity ; GO:0097063 cadmium ion regulated ... TF activity involved in negative regulation of transcription ; GO:0001220

Original comment by: rebeccafoulger

[gocentral]

Hi,

I'll add a couple more comments to this item. My original plan was to create new transcription factor terms for the copper and cadmium ion sensor txn factors, and then obsolete the original "copper ion sensor activity" and "cadmium ion sensor activity" terms as discussed and agreed previously in this SF item.

I have already made the new txn factor terms, which are in the txn factor activity tree and which have has_part relationships to the appropriate metal binding terms. However, I did not proceed with the obsoletion because Paul Thomas had some concerns which he, David, and I have been discussing via email. So, at this point, there are two terms present in GO, both intended for the same metal binding txn factors. One way or the other, they should be consolidated.

David, in response to your question about making "copper ion sensor activity" a 'receptor activity', we discussed this same question for the case of the cerevisiae gene Cup2 and decided that this was not a receptor activity. So, if it is decided that these ones (copper and cadmium ion sensor activities) are receptor activities, then the entire metal-regulated txn factor activity tree should probably be integrated into the receptor activity tree, not just these two terms.

-Karen

Original comment by: krchristie

[gocentral]

The reason why I suggested the receptor route is that retinoic acid receptors/transcription factors seem to work in a similar way as the TF terms here and they are definitely considered receptors (I used to work on them).

I guess another question is that if the 'sensor activity' type terms are not synonymous with with the TF terms, then there should be some gene products that bind the ligand and change a cellular process in a way other than as a TF.

If I remember correctly, the original reason for wanting the obsoletion was because we thought the original terms were vague and they were really meant to be the TF terms.

David

Original comment by: ukemi

[gocentral]

1/ The only current annotations to: cadmium ion sensor activity ; GO:0097063 copper ion sensor activity ; GO:0097077 are to transcriptional regulators (CmtR and csoR, respectively)

Therefore I'm ok with merging these two metal sensor terms into the TF terms Karen has previously created.

2/ Should the sensor terms have 'receptor' parentage? For other sensor terms, I'm for changing the is_a binding to HAS_PART. soluble guanylate cyclase (a second messenger) is annotated to 'oxygen sensor activity & NO sensor activity' . sGC is oftten referred to as an NO receptor in the literature as it receives and responds to NO (fitting our definition of receptor activity ; GO:0004872).

Where do you draw the line between a receptor (the first point of contact for a signal) and an enzyme activity that is regulated by chemicals/metals?

Original comment by: rebeccafoulger

[gocentral]

I think this comes down to the signaling aspect of the term. If the molecular function somehow relays information from the ligand to the cell to regulate some other cellular process,, then we could consider it a receptor.

A simple up-regulation of an enzyme activity wouldn't fit this, would it?

I also think that we have to be consistent. If theings like the NO receptor and the RA receptors behave in exactly the same way as these metal sensors, then I think we are getting ourselves into trouble by saying some are receptors and some aren't.

>Where do you draw the line between a receptor (the first point of contact >for a signal) and an enzyme activity that is regulated by >chemicals/metals?

Original comment by: ukemi

[gocentral]

The receptor terms were split out last year, mainly into signaling receptors (those associated with a downstream signaling cascade) and cargo receptors (e.g. for endocytosis)

I am happy to say that the remaining sensors are 'receptors'. Some can go under 'signaling receptor' I'm sure. They're a bit lost at the moment under the binding node.

Original comment by: rebeccafoulger

[pgaudet]

I am adding label 'MF refactoring' because I think these types of terms are being discussed.

Pascale

[thomaspd]

In the TF refactoring, we should have a term to cover copper ion-regulated transcription factor activity. We don't necessarily need detail about which polymerase, etc. In this way the term can cover all domains of life, as the main point is that it is a compound function where one domain/function activates another domain/function, like nuclear receptors.

[ValWood]

@pgaudet I think you can close this as out of date?