Itaconate modulates immune responses via inhibition of peroxiredoxin 5, 2025, Tomas Paulenda et al

[V.R.T.]

That’s exactly my hunch, but it’s pure speculation at this point so I hope nobody puts too much faith on it until there’s data to back it up.

Unfortunately trying to design and carry out wet lab experiments is not something that can be easily done by someone with ME/CFS who has thus far been completely limited to computational work. Even if you do find a magic supplement that helps a lot. But lord I’m trying!!


I hesitate to speculate on it at this point since the only options I currently see are quite heavy duty drugs and I definitely don’t want to put it out there without evidence. I’d rather have the mechanism be confirmed and jump from there

Well, I think it goes without saying that we all appreciate your efforts! I remember how hard doing academic humanities work was with mild ME, I can't even imagine throwing lab experiements into the mix!

Are the lab experiements ongoing or are they still being planned/set up? Do you have any sort of idea of a timescale until you might have your answer?

It does sound like a plausible theory that would fit with my experience of deterioration, and definitely sounds worth pursuing.

Totally understand about the drugs. I think that's probably sensible given the direness and desperation of our situations.

 

[jnmaciuch]

Are the lab experiements ongoing or are they still being planned/set up? Do you have any sort of idea of a timescale until you might have your answer?

I have a potential collaboration worked out where a generous research team might be able to carry out one preliminary experiment for me. Timeline is a few months at least.

In the mean time, I’ve been meeting with researchers at my university to try to get trained on some protocols. Likely I’d need to get a grant award before I can move forward with doing anything myself. I’ll probably have a better chance with a grant if that preliminary data comes back supporting my idea

 

[RDP]

It also would be great to know about the effect of itaconate on the other PRDXs.

Max Artyomov was asked about the other PRDXs during the Q&A of his recent Harvard talk. He said that he believes other PRDXs respond the same as PRDX5.

 

[RDP]

Does anyone know if Robert Phair has published his idea on an alpha interferon positive feedback cycle?

Hi Jonathan, There is no publication. I'm just obsessed with finding a cure for this damn disease. The hypothesis has two parts: 1) the shunt itself (cis-aconitate-->itaconate-->itaconyl-CoA-->citramalyl-CoA and then a lyase (CLYBL) that brings the C5 pathway to its short circuit end: pyruvate and acetyl-CoA. This first part would explain energy inefficiency in ME because the "itaconate cycle" produces at most one NADH. 2) a chronicity mechanism - some pathway that makes the shunt a long term feature of central carbon metabolism in affected cells. My first guess about this was based on recognizing that Type I interferons are, themselves, ISGs. We measured plasma IFNa in HCs and MEs using the Quanterix Digital ELISA and, disappointingly, found no difference [again, not published, but reported at the Hinxton Invest in ME meeting 2 years ago]. Since you've invested the time to watch those videos on the "Janet Show", you've heard my concern that plasma measurements may be fruitless if the ME/CFS phenotype is restricted to only 10-15% of body cells. That ME/CFS is a cell autonomous disease is a strong inference from Cara Tomas' and Julia Newton's results in fresh or cryopreserved PBMCs showing resting ME/CFS VO2 (Seahorse) decreased by nearly half. If every cell in the body was this sick, then whole body VO2 would also be half normal. Various investigators have measured VO2 for whole human HCs and MEs. The Vermeulens' and David Systrom's are my favorite papers on this. Apparently depending on disease severity, they measure VO2 between 86% and 95% of normal resting VO2. This, of course, is not treated as a major result of those papers, but it should be because I do not see any way around the conclusion that only 10-15% of patient cells have the Tomas/Newton phenotype [again no paper, but reported in my talk for the NIH ME/CFS Roadmap Physiology Webinar]. If one accepts this conclusion, it's but a small extrapolation to the classic ME patient's "normal bloodwork." Blood concentrations of anything are weighted averages of contributions from all body cells. The endocrinologists have it easy because they can measure signals uniquely produced by a small fraction of body cells, but if we think that ME is a disease of central carbon metabolism, plasma measurements will carry very little information if only 10-15% of cells are sick.

For this same reason, I'm constantly worried that I'm not studying sick cells even though I am studying cells obtained from ME/CFS patients. Even at the level of scRNAseq, ME PBMCs are not all different from HC PBMCs. At another meeting, Andrew Grimson agreed with this point: "At the level of mRNA, most PBMCs are normal, not sick." This is why Andrew's PD subset of classical monocytes is so interesting.

Getting back to the positive feedback loop, the reason Tom's and Max's Nature Metabolism paper is so compelling is that it provides yet another positive feedback loop that could explain chronicity of the itaconate shunt. It even explains why we find increased mitochondrial superoxide (MitoSox) in ME mitochondria, something I could not explain as a consequence of the itaconate shunt alone. Professional immunologists like Tom and Max tell me its impossible for any feature of innate immunity to be sustained chronically. If that's true, the itaconate shunt hypothesis is toast. I would counter that in computer simulation, a broken off-switch or a positive feedback loop can make the impossible possible.

On the other hand, experiments are the classic way to turn a hypothesis to toast - who was it who said, "...a beautiful hypothesis destroyed by an ugly fact"? We have an ME PBMC phenotype in which permeabilized ME cells produce significantly fewer reducing equivalents (as measured by a tetrazolium redox dye) from a fixed amount of a TCA cycle fuel. We did this for multiple fuels, and, intriguingly, reducing equivalent production was most impaired when the fuel was cis-aconitate. I was, admittedly, pretty hopeful that this result implied the itaconate shunt was active, but we measured ACOD1 mRNA and CAD protein and neither was increased compared to HC. We're still in the process of measuring itaconate, itaconyl-CoA, and citramalyl-CoA. If we don't find them increased, then the itaconate shunt hypothesis is experimentally toast assuming that a PBMC phenotype is a cellular model of the ME disease mechanism. Vide supra.

 

[RDP]

Evolution of a positive feedback loop that doesn't have a spring loaded return mechanism seems to me implausible.

True. But the evolution of biological feedback control comes at the cost of nonlinearity and nonlinear systems have features that linear systems do not. Prominent among those features is bistability. A simple definition of bistability is that the system can operate in two different steady states for the same set of inputs and parameters. There are solid examples of this in human metabolism. Those of you with long memories will remember that bistability was a key feature of my first hypothesis for the pathogenesis of ME, the IDO metabolic trap. If one of those stable steady states is pathological, evolution will not be aware of it because the same metabolic control system is also capable of normal health. The gene circuit producing the sick phenotype can also produce the healthy phenotype. Indeed, both phenotypes are inherent in exactly the same biochemistry.

This raises the thorny problem of predisposition. What is it that explains why some individuals suffer an infection or trauma that devolves to ME, and others, suffering the same insult, do not? In nonlinear control theory, the bistable nonlinear system needs an exogenous stimulus to transition form one stable state to another. Of course, this could be a case of I've got a hammer (nonlinear control theory) so everything is a nail (bistability). Still, the possibility that ME/CFS is a bistability disease seems worth exploring.

One possibility I find attractive is that predisposition to ME/CFS is acquired during a previous infection (one preceding the "trigger" infection). Many viruses have evolved non-structural proteins that in various ways sabotage the host immune system. By this mechanism, the ME trigger infection finds a host immune system not entirely defined by host evolution. Indeed, the saboteur virus may have delivered a Trojan Horse that disables the homeostatic, spring loaded return mechanism. The viruses have been evolving a lot longer than we have.

 

[Jonathan Edwards]

Hi Jonathan, There is no publication.

Hi Robert @RDP , thanks for re-joining the discussion.
I think this sort of dialogue could be very productive. I don't think anyone has quite got the answer here but people coming from different angles picking things apart might get us there. This is exactly how we picked apart rheumatoid system dynamics and just one new observation (which it turned out we had made ourselves five years earlier and missed) allowed us to fit a whole lot of known steps into an autonomous story. I need to read through your comments carefully and respond. My collaborator, Jo Cambridge, who also worked with Chris Armstrong on this, is closer to your position than I am, but I think we are pretty much agreed on the systems dynamics arguments. It is just working out how the pieces fit together.

 

[Jonathan Edwards]

Professional immunologists like Tom and Max tell me its impossible for any feature of innate immunity to be sustained chronically.

I eventually became a professional immunologist having been a rheumatologist, macrophage and stromal cell biologist, histopathologist, embryologist and fluid physiologist along the way! That might make me a Jack of all trades but I think it taught me to have a group of people with lots of different knowledge bases - redox, antibody, macrophages, tissue mechanics, whatever. (Macrophage FcgammaRIIIa is induced by mechanical forces, which is why RA is an arthritis.) I suspect @jnmaciuch may be picking up the same vibes from a similarly diverse track.

I can see what Tom and Max may be saying, and it is related to my own concerns, but I don't think it is as simple as that. There is no independent innate immune response. Even newborn infants use antibody to forearm their macrophages and NK cells. As Donne said 'No cell is an island, unto itself'.

Coming from a histological and tissue structure training I tend to see everything situated in this context and I don't think we have thought enough about where we think the bad cell events really are in ME/CFS. I am writing something with Jo Cambridge and Jackie Cliff and both of them came back with the first comment - 'where is it?' I think it very plausible that metabolic shifts inside macrophages are crucial to the bad signalling that makes people feel ill but this bad signalling might be hidden away in places we have never looked. I don't think we even need muscle cells to be involved at all, although they may be.

Where I suspect I agree with Tom and Max is in that it is hard to see innate immune cells getting stuck on the wrong fork of a bistable pathway without some outside help. And especially if this problem is supposed to spread to cells all over. I think the evolutionary argument must be that a pathway that is bistable and can flip into a hypo metabolic state entirely without external help will not survive. Non-linear bistable pathways are everywhere in haematology and immunology but they all have switch points that have external control I think.

The counter to this argument would be the pyrin gene allele for familial Mediterranean fever perhaps. It may have advantages as a heterozygous allele but produces bouts of fever when homozygous. It might be that people with early onset ME/CFS have something analogous - a gene variant that allows a purely innate immune cell bistable pathway to stuck flipped. It is just that the dynamics of ME/CFS seem very odd for that. It is not a paroxysmal disorder, nor a purely progressive one. It seems that other forces must be having a variable permissive effect on a flip over very long periods. That looks to me like an 'adaptive' immune response, with the caveat that you can get learnt clonal expansion of T cells that work more on innate signals than antigen recondition.

So I like the basic idea of invoking a positive loop and I like the idea of interferons because they modulate ell behaviour without necessarily being directly pre-inflammatory. I am drawn intuitively to gamma interferon but I have an open mind.

 

[Jonathan Edwards]

One specific thing that I think may be putting people of the scent is that peripheral blood mononuclear cell metabolism probably tells us thing much. A monocyte or a B cell only does something of interest when it meets other cells in tissues - when it may expand its cytoplasm and metabolic machinery tenfold. The Catch22 is that what it does wrong in ME/CFS may depend on soluble factors that may be present in the original plasma but will be lost in standard tissue culture. They may also only be present at short range between interacting cells. So failing to find things maybe should not put us off completely.

One possibility I guess is that, as for lupus, there are people with such a high genetic predisposition to ME/CFS that they are bound to get it and quite likely by age 12. Maybe once we have some genetic clues very early onset cases should be looked at as a specific subgroup. It might be that these people have such a bistable pathway in their macrophages that you can actually study it in vitro in isolation regardless of soluble mediators in the medium.

 

[Sasha]

This is fascinating:

One possibility I guess is that, as for lupus, there are people with such a high genetic predisposition to ME/CFS that they are bound to get it and quite likely by age 12. Maybe once we have some genetic clues very early onset cases should be looked at as a specific subgroup. It might be that these people have such a bistable pathway in their macrophages that you can actually study it in vitro in isolation regardless of soluble mediators in the medium.

I'd never considered that we should focus on early-onset cases. Ron Davis focused on the severely ill to get the biggest ME/CFS 'signal'. I wonder if early onset and severity are correlated and if not, why not.

 

[Utsikt]

Surely there must be some surveys that have data on age at onset and severity? Could we do some as hoc analyses on those?

 

[Sasha]

Surely there must be some surveys that have data on age at onset and severity? Could we do some as hoc analyses on those?

Actually, wouldn't DecodeME have this data for 25k people?

@Simon M

 

[Sasha]

But I suppose there would be all sorts of confounders in a cross-sectional study so we'd need a prospective one.

Anyway, I think I may be derailing the thread. Maybe we should start another one if we want to discuss this particular sub-topic.

 

[Jonathan Edwards]

Actually, wouldn't DecodeME have this data for 25k people?

My thought is that if DecodeME comes up with genetic links it would be very useful to look at patients presenting under 15 to see if the link was enriched. If nothing else it would be a way to strengthen the significance of the data. At leat some genetic links should be enriched I early inset cases but they might be rare allele ones that need while genome screening.

 

[Kitty]

My thought is that if DecodeME comes up with genetic links it would be very useful to look at patients presenting under 15 to see if the link was enriched.

Might there are also be a case for also looking at a group aged 15 – 21 at onset, to see if it suggests there could be a link with late acquired EBV?

 

[Jonathan Edwards]

Might there are also be a case for also looking at a group aged 15 – 21 at onset, to see if it suggests there could be a link with late acquired EBV?

Yes, but I think very early cases are most likely to give a clear signal.

 

[jnmaciuch]

Yes, but I think very early cases are most likely to give a clear signal.

Ha! I woke up with the exact same line of thought based on my earlier comment:

Which would then prompt you to ask why someone with these mutations is not getting ME/CFS at their first viral infection in childhood. The answer would be that whatever mutation exists, its effect is modulated through a variety of other factors, such as sex hormone levels, the effects of chronic poor sleep quality, etc. etc. etc.

It would make sense that more deleterious mutations in the same pathways would end up rare in the population if they severely affect children at earlier stages.

The recent rare variant WGS results might actually be quite helpful in that regard then—it’s why I was thinking proteasome in the first place in my throwaway “boot” theory. Glutathione levels also came up as predictive in Leonard Jason’s prospective EBV cohort.

At risk of flattering myself too much I’ll take that as a “great minds think alike” moment.

 

[paulendat]

I eventually became a professional immunologist having been a rheumatologist, macrophage and stromal cell biologist, histopathologist, embryologist and fluid physiologist along the way! That might make me a Jack of all trades but I think it taught me to have a group of people with lots of different knowledge bases - redox, antibody, macrophages, tissue mechanics, whatever. (Macrophage FcgammaRIIIa is induced by mechanical forces, which is why RA is an arthritis.) I suspect @jnmaciuch may be picking up the same vibes from a similarly diverse track.

I can see what Tom and Max may be saying, and it is related to my own concerns, but I don't think it is as simple as that. There is no independent innate immune response. Even newborn infants use antibody to forearm their macrophages and NK cells. As Donne said 'No cell is an island, unto itself'.

Coming from a histological and tissue structure training I tend to see everything situated in this context and I don't think we have thought enough about where we think the bad cell events really are in ME/CFS. I am writing something with Jo Cambridge and Jackie Cliff and both of them came back with the first comment - 'where is it?' I think it very plausible that metabolic shifts inside macrophages are crucial to the bad signalling that makes people feel ill but this bad signalling might be hidden away in places we have never looked. I don't think we even need muscle cells to be involved at all, although they may be.

Where I suspect I agree with Tom and Max is in that it is hard to see innate immune cells getting stuck on the wrong fork of a bistable pathway without some outside help. And especially if this problem is supposed to spread to cells all over. I think the evolutionary argument must be that a pathway that is bistable and can flip into a hypo metabolic state entirely without external help will not survive. Non-linear bistable pathways are everywhere in haematology and immunology but they all have switch points that have external control I think.

The counter to this argument would be the pyrin gene allele for familial Mediterranean fever perhaps. It may have advantages as a heterozygous allele but produces bouts of fever when homozygous. It might be that people with early onset ME/CFS have something analogous - a gene variant that allows a purely innate immune cell bistable pathway to stuck flipped. It is just that the dynamics of ME/CFS seem very odd for that. It is not a paroxysmal disorder, nor a purely progressive one. It seems that other forces must be having a variable permissive effect on a flip over very long periods. That looks to me like an 'adaptive' immune response, with the caveat that you can get learnt clonal expansion of T cells that work more on innate signals than antigen recondition.

So I like the basic idea of invoking a positive loop and I like the idea of interferons because they modulate ell behaviour without necessarily being directly pre-inflammatory. I am drawn intuitively to gamma interferon but I have an open mind.

Hi Jonathan,

I completely agree, that we can't presume purely myeloid compartment playing a role. This simply doesn't exist in the body. Everything is interconnected. What is really needed now is systemic analysis of immune system. I think what I've seen so far that this analysis is warranted. We need to see how both innate and adaptive immune responses are changing in patients.

The most crucial are timing and location where we look. It is possible that in most severe cases the differences will be profound enough to manifest systemically and can be observed even from PBMCs.

We are currently looking into obtaining patient samples to analyze.

I still believe it is possible for itaconate to play a role. You may say I'm biased and it will be true. It is my favorite molecule. But regardless of itaconate, there are gaps that need to be answered before any conclusions can be drawn.

 

[Jonathan Edwards]

I still believe it is possible for itaconate to play a role. You may say I'm biased and it will be true. It is my favorite molecule.

Don't worry, @paulendat you can easily sell me some shares in itaconatePLC. But yes, we need to know where and when and we need the whole picture. I think now is the time to be looking at this. Things are funelling in to a plausible story. And if my past experience is anything to go by we may have already done the clincher experiment, if only we could see which one it was!

 

[Braganca]

I can’t understand any of it, but this thread is making me so happy.

Thanks to all you smart peeps for trying to figure us out..

 

[Yann04]

Hi Jonathan,

I completely agree, that we can't presume purely myeloid compartment playing a role. This simply doesn't exist in the body. Everything is interconnected. What is really needed now is systemic analysis of immune system. I think what I've seen so far that this analysis is warranted. We need to see how both innate and adaptive immune responses are changing in patients.

The most crucial are timing and location where we look. It is possible that in most severe cases the differences will be profound enough to manifest systemically and can be observed even from PBMCs.

We are currently looking into obtaining patient samples to analyze.

I still believe it is possible for itaconate to play a role. You may say I'm biased and it will be true. It is my favorite molecule. But regardless of itaconate, there are gaps that need to be answered before any conclusions can be drawn.

I love that there is a such a thing as a favourite molecule. To a layman like me it sounds kind of ridiculous but I guess it makes a lot of sense. Anyways, the idea of everyone having a favourite molecule makes me giggle

(if this message reads like insomniac gibberish, that’s because it is…)