NTRs: 'synapse pruning' and child terms

BarbaraCzub commented 6 years ago

I have recently come across several papers describing 'synapse pruning' (e.g. PMIDs: 18083105, 21778362 , 22632727) and its more specific complement-dependent type (e.g. PMIDs:18083105, 22632727), yet, we do not have GO terms describing these processes.

Below are my suggestions:

New1: synapse pruning Definition: Developmental process involving microglial phagocytic engulfment of redundant synapse fragments, required for maintenance of optimal long-term synapse density and activity. PMIDs: 18083105, 21778362 , 22632727. GOC:aruk. GOC:bc. Exact synonym: synaptic pruning Related synonyms: synapse elimination, synapse remodelling, synaptic elimination, synaptic remodelling (or should these be 'broad' synonyms? @paolaroncaglia or @ukemi perhaps you could advise on this?) is_a parents: neuron remodeling; phagocytosis, engulfment

New1a: complement-dependent synaptic pruning Definition: Synaptic pruning mediated by complement system signalling. PMIDs:18083105, 22632727, 29844190. GOC:aruk. GOC:bc. Exact synonym: complement-dependent synaptic pruning

Please let me know, if you have any comments, or suggestions, esp. with regard to the definitions. @Pimmelorus, perhaps you'd be able to comment on this?

Thanks, Barbara

cc: @RLovering @rachhuntley @paolaroncaglia

pvannierop commented 6 years ago

@BarbaraCzub I use the term 'synapse disassembly (GO:0098883)' for synaptic pruning.

BarbaraCzub commented 6 years ago

@Pimmelorus thank you so much for pointing this out! Would you suggest then that 'synapse pruning' should be a synonym of 'synapse disassembly'? This was my initial thought after reading your comment, but having looked at a few papers, I think that perhaps 'synapse disassembly' definition should be broadened and 'synapse pruning' should become a child of it.

For example, the experiments conducted by Eaten, Fetter and Davis (2002; PMID: 12062020) suggest that 'synapse disassembly' could also result from 'synapse retraction'. In a review (PMID: 12952887), which shortly follows this research paper, these same authors also cite studies, which show that synapse disassembly occurs due to loss of signalling involved in synapse maintenance (this seems to fit the sole annotation made for this term so far). But it is not clear whether the loss of this synapse maintenance signalling results in synapse disassembly due to pruning, or retraction.

So, I suppose the ultimate question here is: Does 'synapse disassembly' really ever occur without synapse pruning by microglial cells? @Pimmelorus are you aware of any published experiments other then the paper, revealed by my cursory search, which I am citing above, that show that synapse disassembly could happen without synapse pruning by microglia?

This would affect whether I add 'synapse pruning' as a synonym (and if so, whether it should be an exact synonym), or as a child term of 'synapse disassembly'. As of now, I am leaning towards adding two new child terms to 'synapse disassembly' and broadening its definition.

Atm 'synapse disassembly' is defined as: 'A cellular process that results in the controlled breakdown of synapse. After it starts the process is continuous until the synapse has disappeared'.

My suggestions:

GO:0098883 synapse disassembly Definition: A cellular process that results in the controlled breakdown of a synapse either due to pruning by microglia or retraction . After it starts the process is continuous until the synapse has disappeared. Exact synonyms: synapse elimination, synaptic elimination Related synonyms: synapse remodelling, synaptic remodelling

New1: synapse pruning Definition: Synapse disassembly involving microglial phagocytic engulfment of redundant synapse fragments. PMIDs: 18083105, 21778362 , 22632727. GOC:aruk. GOC:bc. Exact synonym: synaptic pruning Additional parents: is_a phagocytosis, engulfment; part_of neuron remodeling (@paolaroncaglia, I initially thought this would be is_a, but since synapses overlap with just parts of neurons, I am now thinking this should be part_of - I would be grateful for your feedback on this please).

New2: synapse retraction Definition: Synapse disassembly involving a retraction of a synapse in a distal to proximal fashion. PMID: 12062020. GOC:aruk. GOC:bc. Exact synonym: synaptic retraction

I aim to add these new terms by Wednesday next week at the latest. Could you please let me know by then, if you have any more comments or suggestions?

Thank you, Barbara

pvannierop commented 6 years ago

@BarbaraCzub I am in no way an expert in synapse pruning, but I think that your interpretation of the process and its child terms is incorrect. Take a look at this wikipedia text on synapse pruning:

The three models explaining synaptic pruning are axon degeneration, axon retraction, and axon shedding. In all cases, the synapses are formed by a transient axon terminal, and synapse elimination is caused by the axon pruning. Each model offers a different method in which the axon is removed to delete the synapse. In small-scale axon arbor pruning, neural activity is thought to be an important regulator,[citation needed] but the molecular mechanism remains unclear. Hormones and trophic factors are thought to be the main extrinsic factors regulating large-scale stereotyped axon pruning.[6]

I conclude that 'synapse pruning' describes the loss of synapses (see terms 'synapse elimination' and 'delete the synapse' in the text above) and that there are different ways this can be caused including microglial involvement. So I think that the terms are 'synapse disassembly (GO:0098883)' and 'synaptic pruning' should be considered synonyms.

BarbaraCzub commented 6 years ago

Thanks @Pimmelorus for your quick reply and clarifying this. You're certainly much more of an expert than I am, so I am going to follow your suggestion, and add 'synapse pruning' as a synonym of 'synapse disassembly (GO:0098883)'. It did cross my mind that the 'synapse retraction' referred to in PMID: 12062020 (that only paper which I found mentioning this) might have been intended more broadly as a 'neurite retraction' for which we do actually have a GO term ('neuron projection retraction' (GO:0106028)). So, your feedback is reassuring, thank you.

BarbaraCzub commented 6 years ago

To do:

[x] Find out how many annotations to 'synapse disassembly' are expected from the SynGO project
[x] Check the mechanistic bases for these annotations. Would the mechanism for these annotations always be pruning by microglia (i.e. phagocytic engulfment)?
[x] Based on the above should 'synapse disassembly' be obsoleted/changed to 'synapse pruning'? Or should 'synapse pruning' become a new independent term?

BarbaraCzub commented 6 years ago

None of the papers annotated by SynGO use the phrase 'synapse disassembly'; however, they mention 'elimination', 'pruning', 'retraction', 'collapse', 'refinement', or 'remodelling' (analysis details: https://docs.google.com/spreadsheets/d/1x0PeAEHlsyaBaTN7rf_Oe1icrhkGKtTQIlmDeAP35kM/edit?usp=sharing).

Language definitions:

To prune (from Cambridge Dictionary):

to cut off branches from a tree, bush, or plant, especially so that it will grow better in the future;
to reduce something by removing things that are not necessary.

To disassemble (from Cambridge Dictionary):

to separate something into its different parts.

To retract (from Cambridge Dictionary):

to pull something back or in.

To eliminate (from Cambridge Dictionary)

to remove or take away someone or something.

To refine (from Cambridge Dictionary)

to make something pure or improve something, especially by removing unwanted material.

To remodel (from Cambridge Dictionary)

to give a new shape or form to something.

To collapse (from Cambridge Dictionary)

to fall down suddenly because of pressure or having no strength or support (from Cambridge Dictionary);
to fold something into a smaller shape, usually so it can be stored, or (especially of furniture) to fold in this way (from Cambridge Dictionary) (from dictionary.com);
to fall or cave in; crumble suddenly (from dictionary.com);
to break down; come to nothing; fail (from dictionary.com).

I also checked the top 3 definitions in oxforddictionaries.com and dictionary.com and the are in agreement with the definitions from dictionary.cambridge.org

BarbaraCzub commented 6 years ago

Preliminary conclusions: The terms 'disassembly' and/or 'retraction' may be appropriate to describe processes affecting the synaptic cytoskeleton. This would need to be specified in the definition.

Refs.: https://docs.google.com/spreadsheets/d/1wHRDLAUjaPVlEMwnvU_uHEWzsfKiN2GNOfzEydvBtNI/edit?usp=sharing

pvannierop commented 6 years ago

Just to explain where the use of the term disassembly comes from. It is based on the GO pattern "assembly, maintenance, modification, disassembly" to describe the life-cycle of cellular structures. According to these broad classes any process that removes a synapse, should be placed in the disassembly category. The term synaptic pruning is a term used to remove redundant synapses which is most prominent during development. In that term the dictionary entry "to reduce something by removing things that are not necessary" does best describe the process since during development more synapses than needed are initially created and the synapses that are no longer needed are removed by 'synapse pruning'. In that sense pruning is removal/disassembly of synapses that are no longer necessary.

retraction is certainly wrong, since the synapse does not retract. The axon may do this, but the synapse is there or it is not.

BarbaraCzub commented 6 years ago

Thanks for your comments @Pimmelorus. I see your point. So perhaps I misunderstand 'pruning'? Based on the literature, which I've read, my understanding is that 'pruning' always involves phagocytic engulfment by microglia (e.g. PMIDs: 18083105, 21778362, 22632716, 29844190). This is why I originally suggested that the term 'synapse pruning' should be related in the ontology to 'phagocytosis, engulfment' as well as 'neuron remodeling' (a process which encompasses 'pruning' according to the GO definition). But then what should we do about papers like e.g. PMID: 12062020, where the authors do not mention 'pruning', but they use the term 'synaptic retraction'? They describe what happens when the maintenance of synapse structure is disrupted from within (here due to a dynactin mutation), so, yes, this is not an ideal example, because here the annotation would have been to 'maintenance of synapse structure' (GO:0099558 )anyway. But nonetheless the phrase 'synaptic retraction' has been used (but perhaps this should have been 'neuron projection retraction'?). I agree, the term 'synaptic retraction' is not precise, but surely an axonal bouton, or a dendritic spine, forming a synapse, both comprise structural components that could retract? Presumably this would not need to involve phagocytic pruning by microglia (but maybe it could?), it and would certainly be a part of 'neuron projection retraction' (GO:0106028). So perhaps 'pruning' should be categorised as a developmental process (as a child term of 'phagocytosis, engulfment' and 'neuron remodeling'), whereas 'disassembly' could remain where it is as a broader term?

pvannierop commented 6 years ago

Based on the literature, which I've read, my understanding is that 'pruning' always involves phagocytic engulfment by microglia (e.g. PMIDs: 18083105, 21778362, 22632716, 29844190).

Redundant cellular components are either processed by the neurons themselves via autophagy (see PMID:29709573), via autophagy by astrocytes, or may involve microglia.

This is why I originally suggested that the term 'synapse pruning' should be related in the ontology to 'phagocytosis, engulfment' as well as 'neuron remodeling' (a process which encompasses 'pruning' according to the GO definition).

'neuron remodeling' is correct, 'phagocytosis, engulfment' I do not know whether this is correct for all instances of pruning.

But then what should we do about papers like e.g. PMID: 12062020, where the authors do not mention 'pruning', but they use the term 'synaptic retraction'?

I suggest to correct their use of terminology. Neuronal processes retract, synapses do not.

So perhaps 'pruning' should be categorised as a developmental process (as a child term of 'phagocytosis, engulfment' and 'neuron remodeling'), whereas 'disassembly' could remain where it is as a broader term?

Synaptic pruning is not always occurring in development, but also in the mature animal. I remain with my point that synapse pruning is a term that means that certain synapses are removed, whereas other remain. I think that removal is equivalent to disassembly. If this is a big issue for you, you may need to consult an expert in the field.

BarbaraCzub commented 6 years ago

Thanks, Pim, for your feedback and the PMID to this very recent synaptic pruning review. I'll take a look at the sections and cited research papers on neurons and astrocytes in addition to microglia to get a broader understanding.

BarbaraCzub commented 6 years ago

Hi @Pimmelorus, we discussed this issue in the Ontology Editors' Call today again. And we've decided that, based on your expert suggestions and based on evidence from literature, the terms 'synapse pruning' and 'synapse disassembly' should be synonymous.

However, because 'synapse pruning' appears to be used in published research more frequently than 'synapse disassembly' the term 'synapse pruning' should be the primary GO term, whereas 'synapse disassembly' (as well as '...clearance', '...elimination', and '...removal') will be exact synonyms.

@Pimmelorus or @ftwkoopmans, could you please let me know whether you are happy with this change in context of the SynGO annotation tool? (This term is in the synapse slim). Or should I keep 'synapse disassembly' as the syngo_official_label, like you previously requested for https://www.ebi.ac.uk/QuickGO/term/GO:0097471 (https://github.com/geneontology/synapse/issues/194).

cc @ukemi
@RLovering

BarbaraCzub commented 6 years ago

To do:

[x] Add exact synonyms of 'synapse disassembly' to GO term entry, specifically 'synapse/synaptic pruning', '... clearance', '...elimination', and '...removal'.
[x] Make 'synapse pruning' the primary GO term name. Keep 'synapse disassembly' as an exact synonym.
[x] Add the child term 'complement-dependent synapse pruning' (and exact synonyms: 'complement-dependent/mediated synapse/synaptic pruning/clearance/removal/elimination/disassembly').
[x] Follow up on syngo_official_label.

For references use: https://docs.google.com/spreadsheets/d/1wHRDLAUjaPVlEMwnvU_uHEWzsfKiN2GNOfzEydvBtNI/edit#gid=0

BarbaraCzub commented 6 years ago

@Pimmelorus @ftwkoopmans the name of this GO term has now been changed from 'synapse disassembly' to 'synapse pruning', and 'synapse disassembly' has become an exact synonym.

Would you like me to assign the syngo_official_label to its exact synonym 'synapse disassembly'? Or are you happy to have 'synapse pruning' displayed in your annotation tool?

Please note that names of child terms (incl. regulation terms) will also have been updated: https://github.com/geneontology/go-ontology/issues/16106

Thanks, Barbara

pvannierop commented 6 years ago

@BarbaraCzub Please make 'synapse disassembly' the syngo_official_label.

BarbaraCzub commented 6 years ago

Hi @Pimmelorus ok, thanks for confirming this, I'll make 'synapse disassembly' the syngo_official_label.

geneontology / go-ontology

NTRs: 'synapse pruning' and child terms #16003