Genetics: Chromosome 6 BTN2A2 and BTN3A3 (BTN2A1)

[ChronicallyOverIt]

People have been timid about using monoclonals to block T cell/NK responses that protect against infection. But they may have got over that now. A monoclonal to block the gdT interaction would be an interesting test of mechanism at least.

A few OX40L monoclonals are coming to market soon:

Press Release: Sanofi’s amlitelimab met all primary and key secondary endpoints in the COAST 1 phase 3 study in adults and adolescents with atopic dermatitis

Sanofi’s amlitelimab met all primary and key secondary endpoints in the COAST 1 phase 3 study in adults and adolescents with atopic dermatitis Amlitelimab, dosed every four weeks or every 12 weeks,...

www.sanofi.com

"normalize the overactive immune system, without depleting T cells"

Originally for dermatitis, but I believe they think they will be good general immune T-cell suppression, possibly off-label

Wouldn’t the OX40l MABs cover this?

 

[V.R.T.]

Are there any physician researchers who would have experience giving T cell drugs in the way Fluge and Mella have for B cell drugs?

 

[ChronicallyOverIt]

Are there any physician researchers who would have experience giving T cell drugs in the way Fluge and Mella have for B cell drugs?

If we’re talking the OX40L from above no, they are brand new

 

[jnmaciuch]

Say the problem is as you hypothesised above - how would we test for it? Are there advanced PET scans that could detect this kind of pathology? Or is it just a 'collect enough evidence to justify a theraputic experiment' type deal?

Likely the latter. This would be an exceptionally difficult thing to get even indirect evidence for. The genetics hits might be enough to justify a therapeutics trial though. If it is something driven by gdT cells then cyclophosphamide would wipe them out too, and it has pretty good penetrance into brain and even bone marrow iirc. So we might have supporting evidence if that trial has a positive result. Something like an anti-gdTCR mAb would be way preferable to cyclophosphamide so I should think funding would be easier to get if it's pitched to funding bodies as "let us figure out if this other drug that is way less likely to give people cancer works just as well"

Wouldn’t the OX40l MABs cover this?

Not for this specific hypothesis, probably. OX40L is a co-stimulatory ligand that helps alpha-beta T cell activation specifically via professional antigen presenting cells (in the lymphoid organs). There isn't much evidence for it being relevant for gdT cell activation

 

[V.R.T.]

If it is something driven by gdT cells then cyclophosphamide would wipe them out too, and it has pretty good penetrance into brain and even bone marrow iirc. So we might have supporting evidence if that trial has a positive result. Something like an anti-gdTCR mAb would be way preferable to cyclophosphamide so I should think funding would be easier to get if it's pitched to funding bodies as "let us figure out if this other drug that is way less likely to give people cancer works just as well"

Iirc cyclo had a positive phase 2 and responders were better years after follow up. They didnt go ahead with phase 3 because of how toxic it is and pivoted to dara.

 

[V.R.T.]

If we’re talking the OX40L from above no, they are brand new

I was talking generally in that post.

 

[V.R.T.]

This would be an exceptionally difficult thing to get even indirect evidence for. The genetics hits might be enough to justify a therapeutics trial though.

In that case, maybe there is an argument for getting a good quality pilot or small phase 2 (if drug already approved) study set up now.

 

[Jonathan Edwards]

This paper:

https://www.sciencedirect.com/science/article/pii/S2211124721007579#undfig1

seems to suggest that BTN2A2 might be a sort of decoy for BTN2A1 that doesn't do much except perhaps compete with BTN2A1. It is all pretty complicated but one way of looking at it may be that it doesn't matter that much whether the SNP that confers risk for ME/CFS is linked to 2A1 or 2A2 expression since they seem to be involved in the same process, with 2A1 being the active molecule. But 2A2 affects T cell function through what may be a less obvious route than initially thought so maybe there is another story et to be identified. It doesn't seem to engage the gdT receptor directly.

 

[Jonathan Edwards]

One thing that is unclear to me is to what extent the gdT cell systems 'learns' in the sense of providing an adaptive immune response. If the T cell receptor does not recognise specific peptide antigens then we do not expect clonal selection of the sort seen with standard abT cells. Does that mean that the gdT cell system never 'learns new tricks' and therefore cannot be blamed for the onset and persistence of ME/CFS? Or are there ways in which gdT cell expansion can become a fixture of the immune repertoire simply by expanding germline gdT receptor carrying populations?

There are a variety of ways of building models that handle such questions I think, but it raises some questions maybe about other conditions like psoriasis and ankylosing spondylitis.

There is also the interesting fact that ME/CFS incidence seems to tail off in mid life - quite unlike the way many autoimmune diseases go on occurring at the same or increased frequency, but maybe similar to lupus and some of the spondarhtropathies. The relatively low genetic contribution to ME/CFS and the ubiquity of most of the trigger infections suggests that stochastic events are involved, but maybe not related to stochastic generation of antigen-specific B or T cell receptor bearing clones.

I can't quite see how all this is best interpreted but I sense that rabbits are loose somewhere. The pivot is in a place we have not been looking at maybe.

 

[Utsikt]

@Jonathan Edwards this popped up on Google:

A population of proinflammatory T cells coexpresses αβ and γδ T cell receptors in mice and humans - PMC

T cells classically express either αβ or γδ T cell receptors. We have identified T cells that express both pairs of receptors. These hybrid αβ-γδ T cells exhibit a hyperinflammatory and migratory phenotype and act as first responders in infection ...

pmc.ncbi.nlm.nih.gov

T cells classically express either αβ or γδ T cell receptors. We have identified T cells that express both pairs of receptors. These hybrid αβ-γδ T cells exhibit a hyperinflammatory and migratory phenotype and act as first responders in infection and CNS autoimmunity.

I have no idea what it means or if it’s relevant.

 

[Jonathan Edwards]

I have no idea what it means or if it’s relevant.

It is something that I was thking about earlier this afternoon. What if some T cells are dependent on BTN2A1 signalling and have clonal antigen specificity. As always, the compartmentalisation of immunology into innate and adaptive breaks down. It seems that these cells are rare but that may be a reason why nobody has taken them into account.

 

[V.R.T.]

@Jonathan Edwards this popped up on Google:

A population of proinflammatory T cells coexpresses αβ and γδ T cell receptors in mice and humans - PMC

T cells classically express either αβ or γδ T cell receptors. We have identified T cells that express both pairs of receptors. These hybrid αβ-γδ T cells exhibit a hyperinflammatory and migratory phenotype and act as first responders in infection ...

pmc.ncbi.nlm.nih.gov

I have no idea what it means or if it’s relevant.

The fact that these are involved in CNS autoimmunity is an interesting connection

 

[forestglip]

One thing that is unclear to me is to what extent the gdT cell systems 'learns' in the sense of providing an adaptive immune response. If the T cell receptor does not recognise specific peptide antigens then we do not expect clonal selection of the sort seen with standard abT cells.

Can gdT cells "teach" abT cells or B cells? Like through recruiting them to come see the antigen they detect, leading to clones of the other cells?

 

[Jonathan Edwards]

The reference to CNS autoimmunity probably relates to mouse EAE models which may not have any relevance to human disease. Nevertheless, it is worth noting.

These hybrid cells may not have the right gdT receptors for engaging BTNs. This is all a bit compllcated for me.

 

[Jonathan Edwards]

Can gdT cells "teach" abT cells or B cells? Like through recruiting them to come see the antigen they detect, leading to clones of the other cells?

So that would be a way out. It would need some antigen-specific abT receptor clones but these might recognise some boring ubiquitous antigen that we all have hanging around so you effectively end up with two signalling populations that we all have anyway getting into a sustained meaningless argument. From a systems dynamics point of view there do seem possibilities here but I don't have any grasp of what might be reasonably plausible. And tyhis sort of analysis is way outside what you are likely to find in any review article out there - which is likely to stick to tropes like molecular mimicry.

 

[jnmaciuch]

One thing that is unclear to me is to what extent the gdT cell systems 'learns' in the sense of providing an adaptive immune response. If the T cell receptor does not recognise specific peptide antigens then we do not expect clonal selection of the sort seen with standard abT cells. Does that mean that the gdT cell system never 'learns new tricks' and therefore cannot be blamed for the onset and persistence of ME/CFS? Or are there ways in which gdT cell expansion can become a fixture of the immune repertoire simply by expanding germline gdT receptor carrying populations?

There is definitely “clonal expansion” downstream of gdTCR activation using all the same pathways as in the alpha betas, it seems. Just a much smaller repertoire set compared to alpha-betas. So there could be some repertoire skewing.

The issue is squaring gdT cells being activated somewhere with a lack of evidence of clonal expansion. During something like infection, the expansion of gdT cells is pretty obvious. It should have come up in TCR repertoire or even scRNA-seq studies (specific g/d chains would come up as differentially expressed mRNA). So if this population exists, it would have to either not expand that much or not enter the circulation.

Activated gdT cells also take on the whole array of possible T cell phenotypes, including cytotoxic. So it would have to be a gdT cell activation that specifically doesn’t lead to a cytotoxic phenotype, or else we’re be seeing tissue damage.

All in all it’s quite a narrow range of possibility

 

[V.R.T.]

Again, the gdT conundrum and the broader T cell hypothesis all seems quite similar to the B cell/autoantibody situation, where there are a lot of theories and a lot of things that you'd expect to see that we don't, and a lot of possibilities that would be hard to prove through the basic science.

With the B cells there has been ritux, dara, cyclo and there will be isa and uplinza soon. It seems like there should be equivalent trials exploring T cell hypotheses...

 

[Jonathan Edwards]

There is definitely “clonal expansion” downstream of gdTCR activation using all the same pathways as in the alpha betas, it seems. Just a much smaller repertoire set compared to alpha-betas. So there could be some repertoire skewing.

OK, so that makes sense and should make things easier. If the repertoire is very small then it should be much easier to get data on relevant numbers, rather than chasing every possible abTCR combination clone.

But that raises a question to me. We have good evidence implicating T cells in conditions like psoriasis, ank spond and Reiter's. Onset of ank spond may be determined by developmental and maturation changes in tissues but Reiter's comes after infections. I don't think anyone has successfully identified clonal expansions that can be blamed for the clinical lesions. Maybe that is because finding abTCR clones is like looking for a needle in a haystack where for gdTCR populations it should be much easier.

But if the skewed populations are somewhere like gut-associated lymphoid tissue, feeding signals into afferent nerves I think it might be very hard to identify expansions in even broad gdTCR populations. Circulating cell numbers in blood have never been very helpful, certainly in immune rheumatic disease.