The itaconate shunt hypothesis

Yes, but if you do NMR on a whole leg you have plenty of metabolite! You can do NMR on arms and legs without disturbing the patient at all - other than asking them to put a leg in a ring magnet.
But it has to be done at the correct time. The example used in my system's modelling class was medium chain fatty acid dehydrogenase deficiency, where the patients had to be fasting for the peculiarities to show up in the blood. If the patients had presented well fed the concentration of the problematic fatty acids would have been fine as the body was able to deal with the dehydrogenase deficiency as long as the system is not stressed. Children used to die from this when sleeping as an overnight fast, especially if coupled with a fever or another type of stressor, became too much.

This MCAD disorder is caused by a mutation in the dehydrogenase gene, but adults have been found with the same mutation and no apparent illness (though I don't want to say they are perfectly healthy, we simply don't know anything but the fact that they have survived having a specific mutation that have killed others).

Edit: Adults can present with fatty acid oxidation disorders even if they have been healthy up until that point, so who knows if they continue to go through life without this illness? What mechanism is making them have a higher capacity for fasting/other stressors than others?
 
Last edited:
Well, presumably for ME when the person is feeling terrible would be the right time?
That is if we are trying to blame feeling terrible on an inability to access energy.
Even if it isn't about access to energy something like a CPET with measurements before and at a few timepoints after could be a good start.
Though in my own illness I don't have to feel terrible to be without enough energy to walk across the living room. My body just won't work ;) At times this "body won't work" also happens together with flu like symptoms, nausea, some wird itching/pain like sensation, but not always.
 
I would just like to add that I watched an Ian Lipkin and Oliver Fiehn Lecture included as part of the "Metabolomics Consortium Collaborative Webinar Series" in April 2021. Towards the end of the presentation he presented one slide speculating that the GABA shunt was active in the TCA cycle of people with ME/CFS. In Robert Phair's Itaconate video he also speculates the GABA shunt is active.

Lipkin5.jpg

I believe that you can still register to watch the recording here
https://ufl.zoom.us/webinar/register/WN_GYSwqAdKT2i1qBDkDGWZwA


On another note, in Ian Lipkins latest metabolomic paper the abstract mentions "Elevated levels of dicarboxylic acids were consistent with abnormalities in the tricarboxylic acid cycle" which is yet another datapoint indicating the TCA cycle should be looked at in more detail, so I am grateful this work is being carried out, with funding from Vinod Khosla.

Evidence for Peroxisomal Dysfunction and Dysregulation of the CDP-Choline Pathway in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome, 2022, Lipkin
https://www.s4me.info/threads/evide...s-chronic-fatigue-syndrome-2022-lipkin.24179/
 
Last edited:
Is this the sort of hypothesis that the DecodeME GWAS study could rule in or out?
I think so - if there is a pathway that gets out of balance in ME (the hypothesis), you would expect to see sDNA differences in genes involved in the pathway or in its regulation.

From my dodgy memory banks, a quick summary of the NMR vs metabolomics approaches

Like all 'omics, metabolomics is a mass discovery approach - looking for everything. It uses mass spectroscopy to find peaks of metabolites, even though it often can't identify what the peak represents.

NMR is different: it's a very targeted approach (or was in the 80s when I came across it). It uses spin-labelling so that the molecule(s) of interest show up easily, much like using a radioactive tracer. so, you first identify which metabolites you want to study, spin (or isotope) label molecules (e.g precursors that will then show up in the metabolite of interest) and look to see what happens. It's very specific.

ADDED And because you are labelling a compound (rather than trying to detect it native stuff), it's much easier to detect. All the unlabelled stuff disappears and the labelled compounds - showing you what you want to see - light up.

As JE put it:
It may be that to get results relating to effects of exercise you need to make use of glucose labelled with a carbon isotope to follow the metabolism through. I forget.
 
Last edited:
Not adding much but comments above ring true from very limited experience.

I recall a PhD biochemistry student who had cobalt/B12 deficient sheep and control (not deficient). The ruminant samples, from the deficient animals, had a huge peak barely present in the controls. The student, of course, identified the difference and had it investigated - succinate which was converted to propionate (by B12) in the control animals (not deficient in B12). So the observant student could identify which samples were going through the system, without looking at ids.

Severe B12 deficiency also resulted in methyl malonic acid accumulating.

So, 40 odd years ago, biochemists were doing a decent job with less sensitive tools. Suggests that those highlighting above, that if there's a major switch in metabolism, then something stands out - something accumulates/disappears.

I actually hope that post GWAS (Chris Ponting's genetic study) there'll be targets to focus in on - the kit MS, MRI ++++ has got a lot better and if you know where to look then it may be possible to identify a biomarker.

I'm not attacking good scientists like Robert Phair, and Chris Armstrong, who are attempting to understand ME/CFS. However, I think post GWAS we might be able to identify where to look and what for.
 
Very good to hear there are more studies in the pipeline.

Am I right in thinking that other metabolomics studies (Lipkin, Hansen, maybe Unutmaz, Naviaux) didn't find increased use of glutamine/ate?

Hey Simon, really good point. Sorry for the extreme delay in seeing this.

Yep, true. I am increasingly thinking that the way to approach these wide-net -omics studies (for things that aren't DNA - ie: metabolomics, proteomics) is to focus on them as exploratory grounds for hypothesis generation within the biological and technical context of the study rather than worry too much about how specific things overlap between studies. This is because there will be too many differences in sample type and acquisition technique that will cause some metabolites (for example, or proteins) to not appear, to appear in varying levels, to be hard to distinguish, to be more or less physiologically relevant for the given type of sample, etc. I'm thinking that there's just too great a difference between tissue types for much useful direct comparison, let alone between cells or biofluids, and NMR vs MS, etc. Probably less of an issue for genomics? You'd know better than I on this front!
 
Last edited:
So, the guts of it is that there is an infection, the presence of the pathogen is recognised by toll-like receptors in the cell, causing the infected cells to sending out signals.

There are inflammatory molecules e.g. TNF-a that increase the permeability of the vasculature so that immune system cells can get out of the blood into the infected tissue. Chemokines move into the blood to signal the immune system cells to come to the infection. The marker CD80 moves to the plasma membrane of the infected cell, signalling that the cell is infected and should be killed.

(21 minute mark). Also the infected cells to secrete IFNa, which tells neighbouring cells to prepare - to move into an 'anti-viral state' (it's really an anti-pathogen state).

(24 minute mark) The IFNa makes the neighbouring cells turn on many Interferon-stimulated genes (of which there are at least a hundred). One of these turns on the Itaconate shunt, in order to reduce the energy resources that might be hijacked by a pathogen to replicate. Another one is for the production of more IFNa - so the idea is that this is a loop.

Normally the infected cells will be eliminated, and not be secreting IFNa, and so the neighbouring cells will stop being in the anti-viral state. The idea is that ME/CFS is that the IFNa production doesn't stop, and so those cells never leave the anti-viral state.
 
So, you might think that it should be possible to find more IFNa in the blood of people with ME/CFS. But yes, I don't think you are kinda dumb belbyr, the preliminary results aren't very convincing at all. The median is higher in the ME/CFS samples, but there is an enormous amount of overlap.

31 minute mark

Screen Shot 2023-01-24 at 10.52.15 am.png

Robert suggests that IFNa might be being produced in blood cells by some people, including the healthy controls. Whereas, the IFNa in people with ME/CFS may be high in tissues. He suggests that it's like trying to measure a pollutant in the sea - you might find low levels that aren't much of a problem by themselves, but the problem might be coming from one particular river where the levels are really toxic.
 
Last edited:
So, the guts of it is that there is an infection, the presence of the pathogen is recognised by toll-like receptors in the cell, causing the infected cells to sending out signals.

There are inflammatory molecules e.g. TNF-a that increase the permeability of the vasculature so that immune system cells can get out of the blood into the infected tissue. Chemokines move into the blood to signal the immune system cells to come to the infection. The marker CD80 moves to the plasma membrane of the infected cell, signalling that the cell is infected and should be killed.

(21 minute mark). Also the infected cells to secrete IFNa, which tells neighbouring cells to prepare - to move into an 'anti-viral state' (it's really an anti-pathogen state).

(24 minute mark) The IFNa makes the neighbouring cells turn on many Interferon-stimulated genes (of which there are at least a hundred). One of these turns on the Itaconate shunt, in order to reduce the energy resources that might be hijacked by a pathogen to replicate. Another one is for the production of more IFNa - so the idea is that this is a loop.

Normally the infected cells will be eliminated, and not be secreting IFNa, and so the neighbouring cells will stop being in the anti-viral state. The idea is that ME/CFS is that the IFNa production doesn't stop, and so those cells never leave the anti-viral state.

Hey Hutan (and everyone),

This is almost perfectly correct bar one thing (and to be fair it is not explicitly mentioned)-the Itaconate Shunt hypothesis here is suggested to affect non-infected cells.

The hypothesis suggests that the infected cells are likely to be cleared by the adaptive immune system-but there is still the IFNa secretion to the extracellular space from the neighbouring, non-infected cells which creates a cycle. Those cells also have the Itaconate Shunt present as a result of ACOD1 being turned on and CAD—the start of the Itaconate Shunt/metabolic ramification—all a result of continued IFNa secretion through the associated pathways.

Just wanted to clarify. However, as detailed in the tweet-thread replies I did, there are mechanisms whereby infection-reactivation could still be present and contribute. Though for the basic hypothesis, it is not thought of like that.

It all needs to be tested.

Hope that makes some sense,


B
 
Last edited:
Thanks Ben H, and for your tweets.

Yeah, sorry if that wasn't clear, if I'm understanding you correctly, that was what I meant to convey (and it is clear in the video) -

In the anti-viral state, the neighbouring cells are induced to turn on the Itaconate shut and produce their own IFNa.
 
Thanks Ben H, and for your tweets.

Yeah, sorry if that wasn't clear, if I'm understanding you correctly, that was what I meant to convey (and it is clear in the video) -

In the anti-viral state, the neighbouring cells are induced to turn on the Itaconate shut and produce their own IFNa.

Yep, that’s it! With the infected cells likely ‘gone’—per the hypothesis—thanks to the adaptive immune system.


With regards to the preliminary IFNa results in the plasma, as it says in the video they are looking into why this may be/some of the potential reasons for the outliers etc. Needs looking into for sure.


B
 
Last edited:
So things they are working on now:

1. Checking if the two healthy control outliers are producing IFNa in their blood cells which might be confounding the results
2. (39 minute mark) Checking if the pathways that are supposed to turn off the IFNa feedback loop e.g. SOCS3 are actually doing that in ME/CFS cells. (SOCS stands for Supressor of Cytokine Signalling). also SOCS1, ZBED2, USP18, phosphates, IL10...

There are drugs that can block IFNa, but Robert notes that they need to get some proof to support this hypothesis. IFNa is important, and blocking it is not a good plan unless you really know what you are doing and have a good reason to be doing it. I wonder if there are some people with ME/CFS who have been given IFN-blocking treatment.

Then there's the opposite - what impacts do IFNa treatments have on people? I think people treated with IFNa for hepatitis often suffered side effects that are reminiscent of ME/CFS. I think Professor Lloyd (of the Dubbo study) did some work looking at that .


I think there are different sorts of IFNa - I wonder if that could be playing a part in the variable results in the blood. That wouldn't explain why some of the most severe patients in their sample had virtually no IFNa of any sort. There is the possibility that the hypothesis might be correct for only some people who are diagnosed with ME/CFS.
 
Then there's the opposite - what impacts do IFNa treatments have on people? I think people treated with IFNa for hepatitis often suffered side effects that are reminiscent of ME/CFS. I think Professor Lloyd (of the Dubbo study) did some work looking at that.

Not the Lloyd paper, but see thread: Persistent fatigue induced by interferon-alpha: A novel, inflammation-based, proxy model of Chronic Fatigue Syndrome (2018) where authors including Chalder looking at these non-ME/CFS patients and concluded —

findings from this study support the hypothesis that abnormal immune mechanisms are important in CFS, but only early in the course of the illness, around the time of the trigger, rather than when the syndrome is established.

Commentary at https://www.virology.ws/2018/12/19/trial-by-error-the-new-interferon-cfs-study/
 
Back
Top Bottom