Hyper-reactivity of CD8+ T cells and high expression of IL-3 correlates with occurrence and severity of Long-COVID, 2025, Renner et al.

[SNT Gatchaman]

Hyper-reactivity of CD8+ T cells and high expression of IL-3 correlates with occurrence and severity of Long-COVID
Renner; Stauffenberg; Paulus; Neumayer; Winter-Köhler; Buchtler; Schmalenberger; Blaas; Mohr; Pfeifer; Malfertheiner; Loew; Sester; Bals; Peterhoff; Schmidt; Mack

Following SARS-CoV-2 infection, some individuals develop Long-COVID-syndrome lasting for more than 3 months.

We analyzed blood samples from patients with Long-COVID, controls without persistent symptoms following SARS-CoV-2-infection and non-infected donors without a history of infection.

Long-COVID patients showed clear signs of T cell hyper-activation predominantly in the CD8+ T cell subset with a 4-fold higher expression of CD25 and 2-fold more effector-memory T cells. Following polyclonal T cell stimulation, we found a 2-fold stronger upregulation of CD25 and a 7-fold higher release of IL-3 in Long-COVID. Intracellular staining revealed 5-fold more IL-3-expressing CD8+ T cells in Long-COVID, while GM-CSF, IFN-γ and IL-2 were much less upregulated. These changes correlated with the severity of Long-COVID and persisted for up to 18 months after infection.

Our data reveal a pronounced and long-lasting CD8+ T cell hyper-activation and hyper-reactivity in Long-COVID and speak for a trial of T cell-immunosuppression in patients with Long-COVID.

Link (Clinical Immunology) [Open Access]

 

[Jonathan Edwards]

That would make sense to me. I think we need to see more ELISpot/FluoroSpot-type studies to pick up local cell interaction, where I suspect cytokine shifts will be much more obvious.

I think maybe people have been sidetracked by the 'exhaustion' idea. I think it more likely that CD8s are overexcited. But I haven't had time to go through the data here.

 

[Sasha]

Our data reveal a pronounced and long-lasting CD8+ T cell hyper-activation and hyper-reactivity in Long-COVID and speak for a trial of T cell-immunosuppression in patients with Long-COVID.

What sort of drugs would be involved in T cell immunosuppression, and how horrible are they?

 

[butter.]

That would make sense to me. I think we need to see more ELISpot/FluoroSpot-type studies to pick up local cell interaction, where I suspect cytokine shifts will be much more obvious.

I think maybe people have been sidetracked by the 'exhaustion' idea. I think it more likely that CD8s are overexcited. But I haven't had time to go through the data here.

Would it still be a 'non inflammatory condition' then?

It would maybe not be surprising if the same patients (patients were recruited in 2021 and 2022) would show an 'exhausted phenotype' or at least some significant changes by now?

Fascinating how many seem to assume these conditions are static.

 

[Jonathan Edwards]

Would it still be a 'non inflammatory condition' then?

It depends on what signals are being produced. IL3 is not primarily an inflammatory signal:

Interleukin-3 (IL-3) is a cytokine that plays a crucial role in the development and function of the immune system and hematopoietic system. It is produced by a variety of cells, including T cells, NK cells, mast cells, and other immune and non-immune cells. IL-3 acts on its receptor, IL-3R, to stimulate the growth and differentiation of various cell types, including hematopoietic progenitors, macrophages, mast cells, and megakaryocytes.

 

[butter.]

It depends on what signals are being produced. IL3 is not primarily an inflammatory signal:

Interleukin-3 (IL-3) is a cytokine that plays a crucial role in the development and function of the immune system and hematopoietic system. It is produced by a variety of cells, including T cells, NK cells, mast cells, and other immune and non-immune cells. IL-3 acts on its receptor, IL-3R, to stimulate the growth and differentiation of various cell types, including hematopoietic progenitors, macrophages, mast cells, and megakaryocytes.

This is certainly an interesting finding, I think time course of the disease will be relevant, though. We need to find better ways to stratify ME/CFS patients in the future.

Quoting the study re IL-3:

Among the cytokines measured, the differences between controls and Long-COVID patients were most pronounced for IL-3 as seen by the release of IL-3 in the supernatant and by intracellular cytokine staining of CD8+ T cells. For IFN-γ the data were somewhat inconsistent as there was a somewhat decreased release of IFN-γ and a somewhat increased frequency of IFN-γ+ CD8+ T cells. In recent years, IL-3 was found important not only for helminth and parasite infection [63,64] but also for development of autoimmunity, inflammation and fibrosis. This has been shown in a number of animal models as well as in patients [[27], [28], [29], [30], [31], [32], [33]]. The link of IL-3 with fibrosis fits to the presence of restrictive changes in lung function tests. Apart from organ specific manifestations, which are not present in Long-COVID, many of these autoimmune diseases are associated with general clinical symptoms like fatigue, stress dyspnea and neuro-psychological symptoms suggesting that chronic T cell activation and T cell hyper-reactivity with release of cytokines and downstream effects on innate immune cells could be responsible for general, organ-independent symptoms of autoimmunity and Long-COVID [65]. Of interest, in a clinical study 87 % of patients treated with recombinant human IL-3 developed fatigue as major side effect [34].

 

[hotblack]

When people talk of ‘hyper-activation and hyper-reactivity’ what do they mean/what is the cause? More receptors on cells, more of whatever signalling chemicals being produced, something else, all of the above?

 

[Jonathan Edwards]

In recent years, IL-3 was found important not only for helminth and parasite infection [63,64] but also for development of autoimmunity, inflammation and fibrosis.

Trouble is there will always be some IL-3 or IL-49 enthusiast who claims it is 'important' for autoimmunity and inflammation. Once you start making sweeping statements like this you are in handwavingland. Animal models can be forgotten - you can get them to do whatever you like. I have not heard of any implication of IL-3 in inflammatory disease in my are.

But it might certainly contribute to inflammatory disease in some contexts and a silent non-inflammatory adverse signalling response, in ME/CFS.

When people talk of ‘hyper-activation and hyper-reactivity’ what do they mean/what is the cause?

Good question. Which is why it would be so much better if authors just gave the data in abstracts, not the PR gloss.

 

[V.R.T.]

That would make sense to me. I think we need to see more ELISpot/FluoroSpot-type studies to pick up local cell interaction, where I suspect cytokine shifts will be much more obvious.

I think maybe people have been sidetracked by the 'exhaustion' idea. I think it more likely that CD8s are overexcited. But I haven't had time to go through the data here.

This is not the first time youve mentioned elispot/fluorospot!

Is this this something that can be set relatively cheaply, easily and speedily or is this kind of experiement lengthy, costly and expensive?

 

[Jaybee00]

What sort of drugs would be involved in T cell immunosuppression, and how horrible are they?

JE has brought up this one before (also known as Campath)—and there are some serious SEs.
https://en.wikipedia.org/wiki/Alemtuzumab

There is an overview here with a table.

https://pmc.ncbi.nlm.nih.gov/articles/PMC8987166/

 

[Jonathan Edwards]

JE has brought up this one before (also known as Campath)—and there are some serious SEs.
https://en.wikipedia.org/wiki/Alemtuzumab

The interesting thing about Campath is that it was given to people with rheumatoid, by friends who were still stuck in the T cell dark ages for that disease, in the hope that RA would vanish. What happened is that all the T cells vanished, a bit like in AIDS except all of them, the RA didn't go away and the patients seemed to be fine otherwise.

T cell depletion has not been explored very extensively because everyone is afraid to, but it may be less problematic than is thought. I have always been interested in specific CD8 depletion but could never get anyone to take it on.

Opportunistic infection is a worry, certainly, but in general it has been rare with depleting agents. Covid is a new issue though.

 

[Jonathan Edwards]

Is this this something that can be set relatively cheaply, easily and speedily or is this kind of experiement lengthy, costly and expensive?

These spot assays require technical skill from the researcher, and finesse with it, but there are plenty of people with those. We did some early work on the side to help some friends wanting to set it up. It doesn't require any very special equipment. Fluorochromes have advanced hugely since then.

 

[V.R.T.]

These spot assays require technical skill from the researcher, and finesse with it, but there are plenty of people with those. We did some early work on the side to help some friends wanting to set it up. It doesn't require any very special equipment. Fluorochromes have advanced hugely since then.

That's good to hear it's not incredibly complex to do. Is it also perhaps the sort of study where recruitment could be done through the DecodeME cohort to speed things up and help with selection?

 

[jnmaciuch]

These spot assays require technical skill from the researcher, and finesse with it, but there are plenty of people with those. We did some early work on the side to help some friends wanting to set it up. It doesn't require any very special equipment. Fluorochromes have advanced hugely since then.

Let me chat with a former rotation PI can’t make promises but it’s a possibility

 

[hotblack]

How broad are these immunosuppressants? I wonder if they’d be taking a sledgehammer to a job better suited to tweezers. If the problem is just some cells or just some signalling, what are the hopes for targeted drugs? Are there any which target just one pathway/cytokine/whatever?

Good question. Which is why it would be so much better if authors just gave the data in abstracts, not the PR gloss.

Oh dear, I was hoping it was another term for me to learn that had some specific meaning.

They seem to be measuring expression of CD25 (a receptor) and concentrations of IL-3 and IFN-γ (cytokines) but also looking at the wider population makeup of T cells? I think… My brain is hitting foggy walls trying to get around what they’re saying. Maybe I’ll have better luck later

 

[Jonathan Edwards]

If the problem is just some cells or just some signalling, what are the hopes for targeted drugs? Are there any which target just one pathway/cytokine/whatever?

For me the key problem is what has the immune system 'learnt' and 'remembers' that makes it keep on producing the wrong response. In RA we hoped to make it forget but it still remembered in the end. For one or two diseases you seem to getting forgetting.

As yet we have no real idea what has been learnt that keeps the response in ME/CFS off kilter. So blocking pathways or removing cell populations en masse is likely to be the first stage - if only to test that we have the right pathway (which is what rituximab did).

The problem with the learnt bit is that the immune system seems to have a very effective way of raising a response using a whole bunch of cell clones with different epitope specificities such that it would take ages to work out exactly what those all were even in one person - let alone two people. So the likely strategy I the long term ought to be to find some sneaky way of telling the immune system to forget the disease signals by conning it into thinking it should do that itself. It is likely to do a thorough job if given the right incentive.

Immunotherapy for bladder cancer does something a bit like this - you tell the body to sort itself out with BCG adjuvant.

 

[jnmaciuch]

So the likely strategy I the long term ought to be to find some sneaky way of telling the immune system to forget the disease signals by conning it into thinking it should do that itself. It is likely to do a thorough job if given the right incentive.

The work on engineered toleragenic APCs seems to be promising for some use cases, though likely a ways away from being usable in humans.

https://www.nature.com/articles/s41551-025-01373-0?fromPaywallRec=false

 

[forestglip]

In RA we hoped to make it forget but it still remembered in the end. For one or two diseases you seem to getting forgetting.

One thing I keep wondering about is why would you expect the B cell memory to go away if you only get the plasma cells? Won't the central memory cells just re-proliferate right afterwards? And is there no good way to take out all the memory cells that isn't very risky?

 

[Jonathan Edwards]

One thing I keep wondering about is why would you expect the B cell memory to go away if you only get the plasma cells? Won't the central memory cells just re-proliferate right afterwards? And is there no good way to take out all the memory cells that isn't very risky?

That is a reasonable concern and certainly our logic for standard autoimmune disease. However, it is conceivable that people accumulate plasma cells that produce high affinity antibody to some past antigen that just happens to react enough with 'everyday junk' from gut or bits of virus to cause trouble. If you clear out the old plasma cells there may be no drive to make antibody with those particular properties if the old antigen is no longer around. With autoimmunity the auto antigen is always still around unless you have destroyed the relevant organ totally.

I think this may be being optimistic but if antibody is involved in ME/CFS it seems to be doing something unusual like this.

 

[V.R.T.]

For me the key problem is what has the immune system 'learnt' and 'remembers' that makes it keep on producing the wrong response. In RA we hoped to make it forget but it still remembered in the end. For one or two diseases you seem to getting forgetting.

As yet we have no real idea what has been learnt that keeps the response in ME/CFS off kilter. So blocking pathways or removing cell populations en masse is likely to be the first stage - if only to test that we have the right pathway (which is what rituximab did).

The problem with the learnt bit is that the immune system seems to have a very effective way of raising a response using a whole bunch of cell clones with different epitope specificities such that it would take ages to work out exactly what those all were even in one person - let alone two people. So the likely strategy I the long term ought to be to find some sneaky way of telling the immune system to forget the disease signals by conning it into thinking it should do that itself. It is likely to do a thorough job if given the right incentive.

Immunotherapy for bladder cancer does something a bit like this - you tell the body to sort itself out with BCG adjuvant.

So the next steps at this stage might be something like:

1.Studies to explore/confirm hypothesis of pathways and cell populations involved

2. Trial drugs that broadly block pathways or remove cell population

3. Studies/hypothesis to narrow down which particular signals are involved

4. Trials for more specific drugs that may help disease memory 'forget' permentanly


Obviously all hypothetical and some of this could happen in tandem.