Longitudinal analysis reveals high prevalence of Epstein-Barr virus associated with multiple sclerosis, Bjornevik et al (2022)

Started one a while ago: https://www.s4me.info/threads/moder...rna-vaccines-for-ebv-and-other-viruses.21078/.

 

It was possible at one time to say that vit D problems could be one reason MS is so common in northern latitudes but there number of children with dark skin born in northern latitudes has vastly increased in the last forty years so you would expect them to be particularly susceptible to MS especially as thye are as likely to get EBV as anyone.

MS is also prevalent among Celtic and Anglo-Saxon immigrants to Australia where they soaked up the sunshine for years before making the connection with skin cancer.

 

I'm not sure this is a real thing vs. the inverse correlation between proximity to the equator and access to modern health care. There aren't many countries between the tropics that have the capacity to deal with diagnosing MS.

It sure would be great to find out, though.

 

It seems to be a real, although you might argue that certain latitudes are preferred by persons with a cultural affinity (and an associated genetic heritage) for living at those latitudes.

Perhaps the reason one doesn't see any countries in the Southern Hemisphere among those with the highest prevalence rates might be because there's relatively little land mass at the higher southern latitudes*. The southern tip of Africa and the southern coast of Australia are pretty much both about as close to the South Pole as Los Angeles is to the North Pole.



https://www.healthline.com/health/multiple-sclerosis/facts-statistics-infographic#Prevalence

ETA: *except for barren Antarctica and the southern tips of Chile and Argentina.

 

(Don't learn geography from that map though, they mixed up the Czech Republic with Hungary. Otherwise, it is of course interesting.)

 

I seem to remember that the Faroes Islands have very high rates. 10% keeps coming to mind but that does seem very high.

I was reading a twitter thread about whether MS was once seen as hysterical paralysis where it was said that MS has always been diagnosed by neurologists even before MRIs were available. The problem with that is MS is easy to diagnose when it affects the patient badly but that is not always the case. MS has been found at autopsy in people who had never been considered ill.

So there could be different explanations. If there is a lot of overt MS then milder cases, particularly in the same family, could be getting diagnosed more frequently in some countries. My friend had a colleague who was obviously sick then her children but it was hard to get a diagnosis. Turned out that despite her Celtic heritage they had Mediterranean familial fever (?) as a random mutation in her genome but doctors did not think to look for it at first.

It is also possible that EBV can make a mild case of MS much worse so it is obvious. My friends with MS talk about strange episodes of dropping things or tripping over their feet or bladder problems for a few weeks decades before they became ill.

Nowadays people are diagnosed much earlier while still appearing healthy without disability, in a third world country there is much less access to MRIs. Then again are more people infected with EBV in infancy in poor countries?

Needs research but MS may get it.

 

Those maps are intriguing. They make you wonder what the rates are in Ireland, and Alaska, and Finland, and the Baltic states, and Russia.

 

From the healthline link I'm amazed that San Marino and Cyprus have such high levels, being so far south. But then San Marino apparently has fewer than 35,000 people living there so perhaps there is a cluster of people with an unfortunate genetic makeup which makes them more likely to get MS.

 

A new paper in Nature says:

Here we demonstrate high-affinity molecular mimicry between the EBV transcription factor EBNA1 and the CNS protein GlialCAM, and provide structural and in-vivo functional evidence for its relevance...
Our results provide a mechanistic link for the association between MS and EBV, and could guide the development of novel MS therapies.

It seems to be unusually good evidence for molecular mimicry, including a mechanism as to how this could cause disease. Maybe @Jonathan Edwards could take a look.

While I'm here, a Science article discusses EBV causation/triggering in the context of the MS/EBV paper. It's a bit heavy going - skip to the graphic to see the relevant bits on causation.

 

DecodeME has tweeted about the relevance of this to ME/CFS. And the MRC retweeted DecodeME...
https://twitter.com/user/status/1485991112393342981

 

Cross-reactivity with individual antibodies is never very surprising, especially when using peptides as antigen. There are a lot of unanswered questions raised by a mimicry theory like this. Mimicry with EBNA1 has been proposed in other contexts and not comet anything. I will wait and see if anyone can confirm.

 

Have you looked at the manuscript? preprint version: https://www.researchsquare.com/article/rs-1239863/v1

I'd appreciate an expert look on this to see whether I'm full of shit or not - while there is some evidence of cross reactivity, the results don't provide a completely convincing story that the induction of GlialCAM antibodies is solely due to molecular mimicry.

They sequenced CSF B-cell receptor sequences from MS patients (and expressed them as antibodies) and found a fab region of one antibody "MS39p2w174" which had a high affinity for GlialCAM and a little bit of cross reactivity with EBNA1.

But the story they present suggests the mimicry is more of a coincidence, rather than a key aspect of the mechanism that led to autoreactivity.

They stated:

Crossreactivity of naïve B cells to self antigens is not unusual, but the key point is that after affinity maturation, the affinity of MS39p2w174 to EBNA1 only increased ~2 fold, whereas the affinity to the post-translational modified GlialCAM increased ~55 fold. Which suggests GlialCAM is the primary target all along.

If the mechanism of inducing self-reactivity to GlialCAM was due to "molecular mimicry", then surely the affinity of MS39p2w174 against EBNA1 would have also increased at a similar rate?

The authors discuss post-translational phosphorylation of the GlialCAM AA370-389 peptide (which is the putative binding region) in specific regions leading to higher affinity binding to MS39p2w17, and indeed it made a big difference in affinity. But this affinity was still much lower than to the GlialCAM protein itself. (also, the KDs in this section are confusingly presented out of order and I think one of the SDs is in error too, given two different SDs were given for the same KD value for the same peptide - hopefully this error will be corrected as I don't think this is the final version of the manuscript)

Next, the putative mimic region between EBNA1 AA386-405 and GlialCAM AA370-389 is only six peptides long and one of the peptides is different! (figure 4j) Six peptides is too short to be compelling.

The primary basis for their claim of molecular mimicry therefore is their EAE mouse model, which utilised the typical incomplete Freunds adjuvant (with mycobacterium tuberculosis) along with EBNA1 AA386-405 peptide or a 'random' peptide as a control. Notably, the experiment induced a strong 11-12 ish fold increase in anti EBNA1 IgG, but a weak anti GlialCAM IgG response ~1.4 fold increase from baseline compared to 1.15 fold increase in the control group. Notably, GlialCAM IgG response increased (1.9 fold change) after induction of EAE with the adjuvant mixture, whereas EBNA1 response decreased slightly over that same time frame. Again, GlialCAM IgG response also rose in the control group (1.4 fold from baseline), though by not as much. It needs to be better demonstrated whether these antibodies were actually pathological.

Lastly, the T-cell assays.

The data shown in extended data 10 k,l do not show a compelling CD4+ T cell response to GlialCAM. This means there isn't a CD4+ T-cell molecular mimicry based mechanism that led to the antibodies against GlialCAM. Now if the high affinity antibodies showed strong cross reactivity against GlialCAM and EBNA1 at the same time, this wouldn't be a big deal. But they don't. While there is some evidence of CD8+ T cell activation in response to GlialCAM, this only occurred in a minority of MS patients - 3 out of the 7 tested.

Also, @Simon M - any comments?

 

Rather than using peptides as antigen perhaps they could produce the protein* or part of it and use that as antigen. OK that all takes time and money.

*e.g. https://www.biorxiv.org/content/10.1101/2021.02.11.430703v1

 

@Snow Leopard Nice analysis.

If I've understood you right, you are saying the authors have shown decent antibody binding of GlialCAM (which could explain MS) but only weak binding against EBV protein EBNA1. And that this could just be because the glialCAM ab happens to bind EBNA weakly - not that EBNA1 is the primary target and glialCAM is caught up via molecular mimicry.

As you say, that affinity maturation* leads to dramatically stronger binding to glialCAM but only a small increase in binding to EBNA1 suggests that the real story is that the immune system is targeting glialCAM and EBNA1 is caught up in things through weak molecular mimicry.

For those new to affinity maturation, it is possibly the coolest thing in human biology, a way to turbocharge antibodies. More in the box below.

Let me know if this is what you were actually arguing.

 

Just to (further) demonstrate my ignorance - if the real story is that the immune system is targeting glialCAM - then perhaps a reliable test for glialCAM autoantibodies = a biomarker for MS.

 

I am afraid I have not had time to read through the full MS. Your comments about differential affinity maturation look cogent. My only caveat would be that if they are basing their proposal a single antibody species forget it. I suspect you can always pull out one antibody that binds quite well to anything. Normally the story is only convincing if the polyclonal response is significant. Admittedly in MS everything is different because you have clones getting into CNS and expanding there so one clone might do a lot of damage.

The basic problem with this sort of mimicry theory is that presumably all of us with EBNA1 antibodies have a whole bunch of clones recognising epitopes including the one supposedly involved. Why haven't we all got MS? What is different about MS is that B cell clones somehow manage to enter the brain. It isn't even clear that it matters what their antigenic specificity is.

The mouse model work to my mind simply makes it more likely that they are not thinking clearly. It is vanishingly unlikely that EAE has anything much to do with the immune dysregulation in MS. There is no reason to think the same antigens are involved. If using EBNA1 makes EAE worse that sounds to me like evidence for what we used to call the yak dung effect. Throw any yak dung into an animal model and it will get worse. (And if it doesn't it doesn't get published that time anyway.)

 

From the pre-print:

Figure 9 is here:
https://assets.researchsquare.com/files/rs-1239863/v1/fb9a8f599bc742551ee2b306.png

It clearly shows that while there is a statistical association, the ELISA assays for anti-GlialCAM or anti-EBNA are not highly sensitive predictors for MS cases compared to controls.

Yes, I'm saying that while there is some weak cross reactivity of the anti-glialCAM antibody against EBNA1, the development of anti-glialCAM IgG seems to be being driven somewhat independently.

This is examining all the evidence together - the weak binding affinity of MS39p2w174 to complete EBNA1 compared to glialCAM, the very short, peptide similarity, the absence of CD4+ T-Cell reactivity against glialCAM, and the weak association in the mouse model.

The key point about affinity maturation is this process is heavily dependent on CD4+ T-cells - either those T-cells are specific for the antigen, or there is an alternative process at play leading to a mistake, such as B-cell cocapture. Also remember that molecular mimicry is usually a cause of tolerance of foreign antigens, rather than autoimmunity due to the regulatory role of T-cells (and negative selection of autoreactive T-cells in the thymus).

 

They expressed a variety of recombinant antibodies from the B-cell receptor sequencing, however few had clear cross reactivity.

Figure 3 m: https://assets.researchsquare.com/files/rs-1239863/v1/92789c3455612437cfe6993e.png

(there was only one clone with clear cross reactivity, though this may be due to incomplete/limited sampling of the B-cells in the CSF)

As far as I'm aware, the oligoclonal banding found in some MS patients is just that, multiple banding rather than a single band?

 

Perhaps no association between MS & anti-GlialCAM antibodies suggests either that: anti-GlialCAM antibodies are irrelevant, or the test for anti-GlialCAM antibodies, used in the study, is crap (poor).