obophenotype
activated vs resting T-cell
@addiehl - relaying a query from HCA curator @zperova:
"While there is a term for the activated state of T cell: https://www.ebi.ac.uk/ols/ontologies/cl/terms?iri=http%3A%2F%2Fpurl.obolibrary.org%2Fobo%2FCL_0001049 there does not seem to be a term for resting T cell
What is the reasoning behind adding a term for activated state but not resting state?"
The CL generally has not attempted to distinguish between resting and activated states of various cell types, except in a few cases where requests from the community for specific activated cell type classes were coupled with clear phenotypic (marker) distinctions between a base state and and activated state that had been shown in the literature.
In many cases for T cells, for instance, the choice of markers used for defining a cell type are true for both resting and activated forms. Obviously for T cells, one could assume a marker like CD69 is generally expressed as a marker of activation, although it is also expressed on some Tregs and memory T cells in a way that may not be a sign of recent activation.
If colleagues working on HCA or related single cell seq efforts would like to propose pairs of resting and activated cell types based on immunophenotyping or gene expression patterns, I would certainly be open to adding activated states of known cell types as classes, if the data is clear and unambiguous.
Marking this as HIPC issue that may warrant additional discussion with the HIPC cell type expert group.
@addiehl thank you for your explanation. This is the paper describing the gene expression profiles of the resting and activated T cells in lung, lymph node, bone marrow and blood: https://www.nature.com/articles/s41467-019-12464-3 Figure 1 shows distinct gene expression profiles for several T cell subpopulations in activated and resting state, while the rest focuses on the differences between the resting and activated CD4+ and resting and activated CD8+ T cells.
Thanks for the paper reference. I will address after discussion with the HIPC cell type expert group, probably in April or May 2020.
Hi @addiehl , Did the HIPC cell type expert group provide any suggestion on this issue, please? As the request came from an HCA query, I'm happy to help implement the changes, if you could provide directions. Thanks! Paola
@addiehl Hi Alex. When you have a chance, could you please look into this? Thanks in advance.
In looking at the paper and thinking about the Cell Ontology, I would like to point out that the T cell classes in the Cell Ontology are not just lexical bins, but include molecular marker information and, in some cases, functional information. When we have added activated T cell classes, these have included additional marker information, such as CD69 or HLA-DR, that are not present on the parent cell type.
Thus, in the absence of molecular information distinguishing a resting T cell from its generic parent or other suitable differentia, data relating to resting T cells should be annotated to the parent, such as 'CD4-positive alpha-beta T cell' (CL:0000624), because the markers on the parent are presumed to be true for all activation states. Data relating to activated T cells can be annotated to a child 'activated T cell' class, such as 'activated CD4-positive, alpha-beta T cell, human' (CL:0001043).
We have only created activated T cell subclasses in rare classes where they were requested and where there was marker information given to differentiate them from their parent. We have one case of a resting cell type in CL, 'resting double-positive thymocyte', which is differentiated from its parent with the logical axiom 'double-positive, alpha-beta thymocyte' and ('bearer of' some 'decreased size').
We could invent PATO classes for 'resting' and 'activated', but that is basically cheating by relying on a circular definition. It's a 'resting T cell' because it bears the quality 'resting'. What is resting, being non-proliferative? Are neurons resting, because they don't proliferate? Are podocytes resting because they have stopped proliferating in their mature state? Resting T cells are not just sitting still, but are often moving through tissues and testing displayed antigens on MHC class I and II.
If we are to create resting subclasses, we need to know by what criteria the authors deemed certain cells as resting. If it is due to the lack of activation markers, then annotating the data to the parental cell types is appropriate. If it is due to the lack of proliferative markers, maybe we can use that as a direct criterion. Looking at the paper, it seems like the authors call T cells resting if they place them in cell culture in the absence of activating anti-CD3/CD28 (which is not a way we can use to define native cell types). Of course, having taken the cells from their natural environment may have already led to phenotypic changes.
I do think there may be some additional classes of T cells that could be added on the basis of this paper, but it would take some study on my part. I am certainly willing to entertain suggestions however. --Alex
@addiehl (cc @zperova and @dosumis) Thank you for the helpful explanation. I'll remove my assignment for now and leave this ticket to Alex then, in case he wishes to add classes of T cells based on the paper suggested by Zina. If not, the issue may be closed.