Reversible reduction in brain myelin content upon marathon running, 2025, Ramos-Cabrer et al.

Reversible reduction in brain myelin content upon marathon running
Ramos-Cabrer, Pedro; Cabrera-Zubizarreta, Alberto; Padro, Daniel; Matute-González, Mario; Rodríguez-Antigüedad, Alfredo; Matute, Carlos

Here we use magnetic resonance imaging to study the impact of marathon running on brain structure in humans.

We show that the signal for myelin water fraction—a surrogate of myelin content—is substantially reduced upon marathon running in specific brain regions involved in motor coordination and sensory and emotional integration, but recovers within two months.

These findings suggest that brain myelin content is temporarily and reversibly diminished by severe exercise, a finding consistent with recent evidence from rodent studies that suggest that myelin lipids may act as glial energy reserves in extreme metabolic conditions.

Link | PDF (Nature Metabolism) [Open Access]

 

That's interesting.

They consider the possibility that dehydration is causing the change in MWF, but they seem confident that it isn't. They mention some other possible reasons for what they are seeing (see below), but still think myelin loss is what is happening.

Could that possibly be what is happening with us? Perhaps somehow certain parts of the brain don't get enough glucose after sustained exertion, and so start using the myelin lipids.

I've seen a marathon runner losing coordination towards the end. I've thought before that it looks a bit like ME/CFS. When I have walked for too long, my coordination goes and movement requires focussed thinking.

At the beginning of my illness, a doctor could not find my knee reflexes despite spending a lot of time looking for them. 6 months later, another doctor found them.

If it was relevant to ME/CFS, it would be downstream of an inability to supply the brain with enough glucose under conditions of mild exertion. Do we have any evidence to support that idea? Could it be some impairment of the TCA cycle arising from a response to an infection (i.e. the itaconate shunt idea?)

But then:
Why hasn't demyelination been seen on brain imaging? (Or perhaps something has been reported?) I guess if the effect is temporary, that makes it harder to find.
I can't recall what has been found with EMG studies of muscle - was the problem with brain signals?

Would it be worth using this technique in a study in an ME/CFS cohort - pre and post exercise?

 

MR imaging of demyelination developed as an assessment of MS (there are other demyelinating conditions, also dysmyelination as a developmental disorder in young children). Demyelinating lesions in that context are discrete and reasonably large. When you see something like "small T2 hyperintensities" we don't know what they are and the differential includes a small focus of demyelination. Here we're talking about a much more diffuse process: larger in area but much smaller in amplitude.

Myelin water fraction imaging is a proxy for myelin content: essentially trying to measure the water contained within the onion skins of the oligodendrocyte-derived myelin sheaths.

Using myelin water imaging to link underlying pathology to clinical function in multiple sclerosis: A scoping review (2022, Multiple Sclerosis and Related Disorders)

So You Want to Image Myelin Using MRI: An Overview and Practical Guide for Myelin Water Imaging (2021, Journal of Magnetic Resonance Imaging)

Magnetic Resonance of Myelin Water: An in vivo Marker for Myelin (2016, Brain Plasticity)

I don't think this particular technique has been used for investigation in ME/CFS. See also this 2023 thread on a preprint, subsequently published —

REUSED: A deep neural network method for rapid whole-brain high-resolution myelin water fraction mapping from extremely under-sampled MRI (2023, Computerized Medical Imaging and Graphics)

 

Thanks for the information and the link to the forum thread SNT.

So, that's sounding pretty interesting as a technique to apply to ME/CFS then?

 

Yes. (Also FND).

Also remember Ron Davis was suggesting demyelination in Catalytic Antibodies May Contribute to Demyelination in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (2023, Biochemistry)

There they were proposing a different cause for demyelination: abzymes / catalytic antibodies. I see I commented at the time —

More diffusely but at much smaller scale, this might be a mechanism though, as summarised by Hutan. (I'll have to look to see if anyone's used this technique in FND already.)

 

The ability to dip into myelin lipid stores as an emergency fuel may have been an evolutionary advantage, allowing our predecessors to cross greater distances and establish new homes (with water, fish etc). So this might be perfectly OK for healthy people with normal (or indeed trained) metabolism.

But what if you're not healthy and metabolism is impaired and you're forced to use more of these structural reserves under what would be normal circumstances? Or there's a group with a genetic flaw that overdoes this process or doesn't recover it swiftly (over 2 months)? They may go beyond subclinical/trivial effects and then show clear impairments of brain functioning.

What if the genes that control this are supposed to ensure it's uniform, but instead it's irregular along neurons and white matter tracts? Then instead of a generalised (let's call it whole-of-brain) slowing of conduction velocities (and therefore brain fog), you might have stochastic / non-deterministic transmission of action potentials (think impaired motor control or non-epileptogenic seizures in FND).

The conventional wisdom is the brain is the one organ without internal fuel reserves, it needs a constant supply of metabolites, oxygen and clearance of metabolic waste or neurons die, quickly. But this suggests that isn't completely true and for a longer scale of more extreme exertion, there is an emergency fuel source that can help - at the potential cost of structural integrity. Remember also, it's been said that metabolically we look like we've run a marathon. This could be an explanation for physical and cognitive-induced PEM.

Why are some of us significantly physically disabled with all the other features of ME/CFS, but not have significant cognitive dysfunction? Do we have better genes that allow for this or do we not have those genes at all? So when we overdo it or have baseline deterioration we're just generally/systemically "cooked" and poorly functioning but still able to think? The descriptions from those severely affected (I think Physics Girl described this) are that they have to simply stop thinking and lie as quietly as possible to avoid worsening. That just thinking makes them worse. Or for those without brain fog, if we do have the same genes, did we lay down more or "better" myelin in infancy (perhaps with fewer early life infections) and maintain that in adulthood, so we can tolerate using myelin lipids temporarily without becoming cognitively symptomatic?

 

Also, this phenomenon doesn't necessarily have to explain PEM, or at least PEM in its entirety. If this is downstream of an inability to use sufficient glucose/operate aerobic respiration on demand, then there are other things upstream and downstream of that might cause PEM.

I note that people who do train physically very hard for extended periods of time do sometimes get something like ME/CFS (over-training syndrome). But, if we were getting this temporary demyelination, why would some people not eventually get enough damage that it becomes permanent demyelination? (Although I know of one person diagnosed with ME/CFS for years who was eventually diagnosed with MS.).

What the idea could maybe do is explain why we seem to temporarily lose good muscle control or cognitive performance after repeated exertion. And yes, as SNT says, perhaps it could explain what is going on with some people diagnosed with FND.

I'm not sure that it all hangs together. But it's intriguing, and it's the sort of thing that could fairly easily and quickly be checked as either something that seems to be a thing or not in a specific patient group.

 

It's interesting to read this paper:
Inactivation of ATG13 stimulates chronic demyelinating pathologies in muscle‑serving nerves and spinal cord, 2025, Drosen et al.
in the context of this finding.

 

The liver stores carbs as glycogen which can be broken back to glucose when needed, as I understand it with a continual flow of surplus glucose from blood to liver stores and back again as needed. And there is fat stored all over the body.

 

We've talked before how researchers need to pay more attention to non-CPET exertion during PEM studies.

It might be that participants need to live close to the testing facility (or there is a mobile testing facility) or the participants go into a residential facility for a week before and after the test.

And there needs to be much clearer instructions about non-CPET activity and monitoring of activity with wearables.

That's an interesting point about the pre-marathon exertion levels. It might be a question for the authors.

 

This process would be just "borrowing" some lipid from the myelin sheath. MS leads to a loss of the oligodendrocytes that produce and maintain the myelin.

Dynamics of oligodendrocyte generation in multiple sclerosis (2019, Nature)

Remyelination by surviving oligodendrocytes is inefficient in the inflamed mammalian cortex (2023, Neuron)

 

Googling, it seems that light sensitivity is common in MS. It is thought that demyelination affects the function of the optic nerve, making the signals go haywire. I wonder if light sensitivity in ME/CFS is due to a more diffuse degradation of the myelin sheath?

Same thing with MS and sound sensitivity. There can be hyperacusis, but also pain in response to sound.