Thread by @ZdenekVrozina

For years, Epstein–Barr virus (EBV) has been linked to multiple sclerosis.
The association was strong.
But the mechanism remained frustratingly abstract - until now.🧵

This new paper in Cell (2026) finally does what’s been missing.
It doesn’t just connect EBV and MS statistically -
it shows how the immune system gets it wrong.

Instead of focusing on antibodies, the authors look at CD4+ T cells.
Long-lived memory cells.
The ones that shape immune behavior over decades.

They find something striking.
More than half of people with MS have T cells that react to ANO2 -
a normal human protein expressed in the central nervous system.

ANO2 is not viral.
It’s not foreign.
It’s part of us.
So why would antiviral T cells target it?

The answer turns out to be EBV.
Specifically, T cells originally trained against EBNA1, a key EBV antigen.
The same T cell clones that recognize EBNA1
also recognize ANO2.

Same receptors. Same clones.
Two very different targets.
This is the turning point.
MS doesn’t look like overactive immunity here.
It looks like misdirected immune memory.

The immune system learns EBV early in life
and carries that memory forever. In some people, that memory is reused -
but aimed at the wrong target.

These are not quiet, harmless memory cells.
They show an activated, cytotoxic profile.
They are ready to act.
EBV infects almost everyone.
Yet only a small fraction develop MS.
So something else must decide who is at risk.

That something else is antigen presentation.
In this study, cross-reactive T cells are enriched in people carrying HLA-DRB1*15:01.
HLA doesn’t cause disease.
It shapes what the immune system learns to see.

If your HLA molecules present EBNA1 peptides
that closely resemble ANO2 peptides,
you create the conditions for molecular mimicry.

When these cells are activated in animal models,
disease becomes more severe.
More inflammation, more relapses, more damage.
Crucially, this pathology is not antibody-driven.
It is driven by persistent, antigen-experienced T cells
that don’t switch off.

Olivia G. Thomas at al., Anoctamin-2-specific T cells link Epstein-Barr virus to multiple sclerosis. cell.com/cell/fulltext/…

Now take a step sideways.
This mechanism is not unique to multiple sclerosis.
In long COVID, EBV reactivation appears again and again.
Often dismissed as incidental.
Possibly it isn’t.

Acute SARS infection creates intense inflammatory and immunological stress.
In that environment, latent viruses - especially EBV - can reawaken. Not in everyone. Not always easy to document. But often enough to matter.

Reactivated EBV means renewed exposure to EBV antigens, including EBNA1.
And that sustains EBV-specific T cell clones!
This study shows why that matters.
Those clones can carry an autoimmune side effect -
molecular mimicry that redirects them toward self-tissue.

Seen this way, SARS-CoV-2 may not be the sole culprit.
It may act as a trigger,
amplifying EBV-focused immune activity in a genetically permissive host.

Add immunological imprinting.
The immune system tends to reuse familiar strategies,
even when circumstances change.
In the MS study, cross-reactive T cells cluster in activated and cytotoxic states.
These are precisely the cells you don’t want making targeting errors.

Translated to long COVID.
Chronic immune stimulation + latent viruses + HLA context
increase the risk that antiviral immunity becomes pathological.
The common thread is persistence.
In MS, EBV persists for life.
In long COVID, multiple sources of chronic immune signaling may coexist.

MS now gives us a clean mechanistic proof.
Other post-infectious conditions may be variations on the same theme.
This doesn’t mean everyone with long COVID has autoimmunity.
It means the terrain is dangerous -
and biology sets the limits.