DNA double-strand break attachment to nuclear envelope versus DNA double-strand break attachment to nuclear cytoskeleton

[lbreuza]

Hi, GO:1990683 DNA double-strand break attachment to nuclear envelope was created on the basis of PMID:24943839 that also described the following Component:

GO:1990612 Sad1-Kms1 LINC complex Cellular Component

Definition A LINC complex implicated in the connection of DNA double strand breaks to the cytoskeleton during DNA double-strand break repair. PMID:24943839 PMID:24947240

Would it make sense to change GO:1990683 to DNA double-strand break attachment to nuclear cytoskeleton. It would then also apply to IFFO1, XRCC4 and LMNA according to PMID:31548606

Alternatively one could change the definition of the Sad1-Kms1 LINC complex which indeed turns out to be a true nuclear membrane complex and create a new process term for IFFO1, XRCC4 and LMNA : DNA double-strand break attachment to nuclear cytoskeleton

[lbreuza]

Maybe anchoring would be better than attachment?

[ValWood]

Hi @lbreuza Do you think these would be better modelled as MF terms?

[ValWood]

I just found GO:0062241 double strand break-nuclear membrane anchor activity Binding to DNA double strand breaks and the nuclear inner membrane, in order to facilitate DNA repair. PMID:31635174

which would cover GO:1990683 DNA double-strand break attachment to nuclear envelope so GO:1990683 could be obsoleted?

Then annotations could be GO:0062241 double strand break-nuclear membrane anchor activity part_of ~DNA repair~ DNA double-strand break repair

I think these 2 processes (nuclear membrane attachment, and cytoskeleton attachment) might be the same thing through different lenses, https://pubmed.ncbi.nlm.nih.gov/29722648/ (and pombe does not have lamins, but I'm sure I saw a paper recently that proposed the bqt proteins (involved in attachment) are functionally similar.

It would be good to be able to align these, isn't the cytoskeleton attachment ultimately about attachment to the inner nuclear membrane? or is there no direct connection? (I don't know much about metazoan IMM)

[ValWood]

Will have a better look Thur/Fri

[lbreuza]

Thanks @ValWood

I just found GO:0062241 double strand break-nuclear membrane anchor activity Binding to DNA double strand breaks and the nuclear inner membrane, in order to facilitate DNA repair. PMID:31635174

I also used that term for the MF of IFFO1 (PMID:31548606) but then wanted to curate a BP which is maybe a bit redundant.

which would cover GO:1990683 DNA double-strand break attachment to nuclear envelope so GO:1990683 could be obsoleted?

Then annotations could be GO:0062241 double strand break-nuclear membrane anchor activity part_of DNA repair

I agree but then you can use double-strand break repair GO:0006302 to be more precise.

I think these 2 processes (nuclear membrane attachment, and cytoskeleton attachment) might be the same thing through different lenses, https://pubmed.ncbi.nlm.nih.gov/29722648/ (and pombe does not have lamins, but I'm sure I saw a paper recently that proposed the bqt proteins (involved in attachment) are functionally similar.

It would be good to be able to align these, isn't the cytoskeleton attachment ultimately about attachment to the inner nuclear membrane? or is there no direct connection? (I don't know much about metazoan IMM) That makes sense if one compare with what is proposed by authors in PMID:31548606. IFFO1 is recruited to DSB in nucleoplasm by XRCC4 and in turns it recruits 'soluble' LMNA which is then integrated into the lamina matrix/cytoskeleton attached with the nuclear envelope. Then LMNA would be functionally similar to the LINC complex. Lamina is attached to INM extrinsically through interaction with inner nuclear membrane proteins like LBR : UniProt/Q14739

[pgaudet]

We should review all 'attachment' terms; these are likely to be MFs.

[ValWood]

I have a list of terms and some nice reviews. This was on my to do list anyways...

These figs are from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6829894/ I'd like to find something equivalent for Metazoa if possible.

Biological processes meiotic attachment of telomeric heterochromatin to spindle pole body meiotic attachment of telomeric heterochromatin to spindle pole body meiotic telomere tethering at nuclear periphery telomere tethering at nuclear periphery negative regulation of mitotic telomere tethering at nuclear periphery meiotic attachment of telomere to nuclear envelope chromosome attachment to the nuclear envelope" (GO:0097240) subnuclear spatial organization of silent mating-type cassette heterochromatin DNA double-strand break attachment to nuclear envelope centromere clustering at the mitotic interphase nuclear envelope attachment of telomeric heterochromatin to nuclear envelope chromosome localization to nuclear envelope involved in homologous chromosome segregation mitotic telomere tethering at nuclear periphery centromere clustering at the mitotic interphase nuclear envelope centromere clustering meiotic centromere clustering

Annotations to the above terms (78) Schizosaccharomyces pombe (44) Saccharomyces cerevisiae S288C (29) Mus musculus (13) Homo sapiens (8) Caenorhabditis elegans (8) Drosophila melanogaster (4) Arabidopsis thaliana

MF EXISTING centromere-nuclear envelope anchor activity GO:0062241 double strand break-nuclear membrane anchor activity

MF suggested

• nuclear envelope-chromosome tethering activity -- telomeric heterochromatin-nuclear envelope tethering activity -- cetromere-nuclear envelope tethering activity -- DNA double-strand break- nuclear envelope tether activity -- telomere-nuclear envelope tethering activity part of "telomere clustering" (if meiotic) -- mating-type region-nuclear envelope tethering activity

Related tickets https://github.com/geneontology/go-ontology/issues/20198 https://github.com/geneontology/go-ontology/issues/18981

[ValWood]

We need to decide which processes should be retained, We could maybe have a term something like

"chromatin organization at the nuclear periphery" then we would use combinations like

telomeric heterochromatin-nuclear envelope tethering activity part_of "chromatin organization at the nuclear periphery" happens_during "mitotic interphase"

I would need to check if we lose anything critical if we migrate to this organization.

[ValWood]

We could also define the specific tethering terms to be precise about whether they connect through to the cytoskeleton or not. Let me know if you come across a nice review for metazoa.

[pgaudet]

Yes for chromatin we need to look into what process this really represents

[ValWood]

Here is the other nice review, also fission yeast , but discusses in the context of human orthlogs throughout. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7464478/

[ValWood]

Looking at DNA double-strand break attachment to nuclear envelope first There are 8 annotations. We need to see if these are a) all directly involved b) could be annotated to the MF term.

They include

• human LMNA (Prelamin-A/C), IFFO1 (Non-homologous end joining factor- Nuclear matrix protein involved in the immobilization of broken DNA ends and the suppression of chromosome translocation during DNA double-strand breaks (DSBs) (these would seem OK to the tethering term)

• SGD MMS21 & SMC6 (SMC complex subunits), RTT107 (Brct domain protein), SIZ1 &NFI1 (SUMO E3 ligases), NFI1 (SUMO E3 ligase)

• PomBase Kms1 (nuclear membrane Linc complex)

The SGD annotations are possibly 'causally upstream' (required for localization, but could be a separate step, because this pathway is probably 'damage recognition' : The pathway is : (some modified?) histone recruits brc1 brc1 recruits nse6 (Smc subunit) SMC complex nse2 (MMS21 in S.c) then sumoylates other Smc subunits (I get a bit hazy here https://www.pombase.org/reference/PMID:30348841 click on the image to embiggen) but possibly sumoylated SMC is what is recognised by the LinC complexes?

@lbreuza which gene product did you originally want to annotate to the attachment BP term?

Even if we do get rid of "DNA double-strand break attachment to nuclear envelope there is clearly a pathway here (DNA damage recognition SUMOylation pathway)

[ValWood]

This seems to be the purpose of the SUMOylation in the DNA damage process https://www.nature.com/articles/srep10984 https://www.cell.com/molecular-cell/pdf/S1097-2765(13)00050-6.pdf

[lbreuza]

@ValWood I curated a GO-CAM model you should be able to check with IFFO1 Hsap. XRCC4, IFFO1 and LMNA collaborate to immobilize ds DNA break in the nuclear cytoskeleton.

[ValWood]

Will do, could you add a link to the model? Thanks v

[lbreuza]

http://noctua.geneontology.org/workbench/noctua-visual-pathway-editor/?model_id=gomodel:63c0ac2b00002148