Genetic association study in myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) identifies several potential risk loci, 2022,Hajdarevic et al

• Largest ME/CFS genetic study to date.
• Three different cohorts totaling more than 2500 patients.
• First Immunochip study in ME/CFS.
• Possible implication of TPPP genetic region

https://www.sciencedirect.com/science/article/pii/S0889159122000782

TPPP plays a key role in myelination of nerve cells and maintenance of the cytoskeleton.

 

Abstract

Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) is a disease of unknown etiology and pathogenesis, which manifests in a variety of symptoms like post-exertional malaise, brain fog, fatigue and pain. Hereditability is suggested by an increased disease risk in relatives, however, genome-wide association studies in ME/CFS have been limited by small sample sizes and broad diagnostic criteria, therefore no established risk loci exist to date.

In this study, we have analyzed three ME/CFS cohorts: a Norwegian discovery cohort (N=427), a Danish replication cohort (N=460) and a replication dataset from the UK biobank (N=2105). To the best of our knowledge, this is the first ME/CFS genome-wide association study of this magnitude incorporating 2532 patients for the genome-wide analyses and 460 patients for a targeted analysis. Even so, we did not find any ME/CFS risk loci displaying genome-wide significance.

In the Norwegian discovery cohort, the TPPP gene region showed the most significant association (rs115523291, P=8.5x10-7), but we could not replicate the top SNP. However, several other SNPs in the TPPP gene identified in the Norwegian discovery cohort showed modest association signals in the self-reported UK biobank CFS cohort, which was also present in the combined analysis of the Norwegian and UK biobank cohorts, TPPP (rs139264145; P= 0.00004). Interestingly, TPPP is expressed in brain tissues, hence it will be interesting to see whether this association with time will be verified in even larger cohorts.

Taken together our study, despite being the largest to date, could not establish any ME/CFS risk loci, but comprises data for future studies to accumulate the power needed to reach genome-wide significance.

 

From the discussion,

"Despite the current lack of large cohorts of phenotypically stringent and well-characterized ME/CFS patients, this study represents, to the best of our knowledge, the largest and most homogenous genetic study performed in ME/CFS so far. Ongoing projects like the DecodeME project (www.decodeme.org.uk) will enable future studies of larger and more powerful cohorts, which are warranted to produce the desired statistical power to definitively investigate the genetic architecture of ME/CFS."

 

Is it a dumb question to ask whether PEM could have something to do with the physical and mental activity stimulating myelin growth and there being limited ability to support this due to some ongoing problem?

 

Not dumb at all. An inability of brain cells to process stimuli with relevant tubulin restructuring would make a lot of sense. If some basic brain process designed to triage and catalogue sensory input were to have its control centre clobbered (probably brainstem) then fatigue, immune activation and all sorts of negative signals might follow.

From what I have seen of the abstract this has the hallmark of creative but modestly presented science that Fluge and Mella produce. The analysis may not prove statistically solid but they are not claiming a definite result - just a finding that if repeated would be very valuable.

 

I'm not sure I understand what that means.

Maybe we are thinking about the same thing? My thought process was: since PEM is delayed and tends to occurs after activity, with sleep inbetween the activity and the worsening, I thought that it would make sense if it PEM is due an ordinary and beneficial response to activity going wrong so that this response ceases to be beneficial or becomes somehow problematic.

Myelin growth seems to be a candidate for such a response, as it's stimulated by mental or physical activity (or more precisely, by electrical impulses in the axons that myelin surrounds).

 

Yes. Sorry about the obscure sentence. I think we are talking about exactly the same idea.

Tubulin biology is very complicated. TPPP is involved in extending tubular chains. Tubulin extension is important for myelin growth and also for nerve cells themselves - and memory and learning. It is likely to be critically important in what happens in your brain during sleep that allows you to wake up the next day with all the household filing done - bills paid put in the bills paid drawer, demands for meter readings put in the 'to do tomorrow when I have time to look under the stairs drawer' and so on. So PEM is the brain saying ' hang on, there is no way you are going out for a stroll in the park while there is this huge mess of things not sorted out from yesterday'.

I remember long ago when I first joined the forums asking whether ME fatigue was more like having stayed up all night or maybe more like having to get up at 4.am to go to the airport. I did that on Sunday and remembered just how ghastly it is. I was told that ME wasn't quite like either but maybe a bit more like the second. Having thought about this so long I had a feeling on Sunday that I might have some idea just how indescribable and disorientating ME symptoms might be - even if they were something else. Basically I had to get up with my brain having been interrupted while trying to catalogue Saturday. Staying up all night is different because the brain hasn't started trying to catalogue so isn't interrupted half way through.

 

Regarding sleep and the brain sorting things out. In the evening I tend to have ideas and plans for the next day, but upon waking up, the details have faded and there is little emotional attachment.

I've mentioned this before as a kind of inability to maintain attention and interest in projects over longer periods (days to weeks).

Maybe this is all just different aspects of the same problem in that the body is too sick to actually do much and plans without realistic ability to carry them out are useless so they get erased during sleep.

 

Can i just say how much i appreciate and respect the way you try to understand us. It just says so much about who you are not just as a doctor but as a person. I'm sure you'll say something like it's 'just being normal' but it's unusual. I mean lots of people try to imagine it, but hardly anyone actually asks "is it like this"? They just assume they know what its like.

So would it explain the sensory sensitivity issues?

 

> whether ME fatigue was more like having stayed up all night or maybe more

to me m.e. fatigue is more like trying to swim from california to sakhalin then to hawaii then back. "it takes a lot out of me." with no sleep. and yakuza following. and accountants with incomprehensible tax forms following. and having to make quick deisions bout killer whales. and auonomic systes failing so breathing rate is weird and a whole lot of other symptoms at the same time. and having to navigate and plan the route without ability to do so. while doing nohing

and tehn the emergency room doctor smirks. and then a deluge of papers says you are a couch potato.

[i really think if we said couch potato instead of deconditioning it would get th absuridty across better.]

> like having to get up at 4.am to go to the airport. I did that on Sunday

i believe sleep can be a strong ediator in the disease. not only because of the asociations [genetic, syptom] with circdian phase but also re malaise and fatigue. closer quaitatively but orders of agnitude different quntitatively.

 

Quote from the study:

 

So... am i interpretting this correctly.... it's something that is 'promising' but needs to be replicated with bigger numbers before getting excited.

And DecodeME should be able to find it if it's there to be found?

 

Are these findings similar to multiple sclerosis?

 

Yes. From the discussion section of the paper, my bolding,

"However, lessons from genome-wide associations in autoimmune and other complex diseases have taught us that several thousand patients and controls are needed to firmly establish risk loci (Kochi, 2016) (Visscher et al., 2017). It is important to stress that we cannot exclude false positive or negative findings in our cohort given our statistical limitation and low allele frequencies of our top hits (Shen and Carlborg, 2013). As has been the history of unraveling the genetic architecture of most common diseases, more GWASs are needed in order to internationally reach the number of patients necessary to provide sufficient power for meta-analyses to establish genetic associations with genome-wide significance.

Despite the current lack of large cohorts of phenotypically stringent and well-characterized ME/CFS patients, this study represents, to the best of our knowledge, the largest and most homogenous genetic study performed in ME/CFS so far. Ongoing projects like the DecodeME project (www.decodeme.org.uk) will enable future studies of larger and more powerful cohorts, which are warranted to produce the desired statistical power to definitively investigate the genetic architecture of ME/CFS.

In conclusion, we identified several potential risk loci for ME/CFS, which encourage further investigations. Future genetic studies should be performed in large cohorts of several thousand patients and strive to use strict and comprehensive phenotyping to enable analyses of homogenous sub-phenotypes."

 

As per comments above, we don't know if this finding will be replicated in a larger GWAS study - DecodeME or whatever. However, on the face of it, this could potentially explain why no biomarker has been identified in ME/CFS i.e.:

• presumably the effect of this gene (TPPP) is to expressing a protein in Cerebro Spinal Fluid (CSF) - not the most accessible. Potentially someone like Jonas Bergquist could re-analyse their CSF samples i.e. looking for this protein in particular; and/or
• the weak correlation is due to the heterogenous nature of the sample population i.e. ME/CFS may be caused by a number of pathologies ---- beginning to doubt that as I type, potentially an alternative is that there may be a number of genes involved in the same process ---- TPPP may not be the only 1.

I think poorly informed speculation probably sums up my comments - to add further evidence - interesting that the signal/correlation seems stronger in the Norwegian samples ---- does that reflect the criteria used to select the two populations and/or different genetics in the two populations?

@Michiel Tack

 

I would say it is intriguing but nothing more as yet - no hard association.
But if DecodeME pulls out the same site it would be something major - even if the DecodeME statistical analysis in isolation is also inconclusive.

If it is spurious DecodeME should draw a complete blank - that is not guaranteed but it is pretty robust, as long as you take into account the number of test points linked to the site of interest. The problem of multiple analyses remains a factor.

 

Yes mutations in the TPPP gene affect multiple sclerosis.

 

A question I’ve never really thought of before is… if you get delayed PEM (next day PEM) does it gradually begin to improve on that day, or does it only begin to improve after another sleep cycle?

I would say the latter.

[I’m not saying it will fully resolve after another period of sleep, just wondering if it won’t start to get better before another period of sleep.]

 

I think its effects may only be found within cells. So yes, this would explain exactly why the mechanism of me is so hard to observe. Tubulin is a polymerising protein that probably does not occur much in fluid compartments and if it does any changes in the TPPP gene may not affect levels. Tubulin is all about solid phase information, not fluid phase.