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

[Jonathan Edwards]

Tom and Jonathan,

Very fair points, Robert @RDP.

Some rogue cell population is hiding in plain sight here I think. We are just not used to the way it is tricking everything else.

 

[RDP]

This certainly fits with other threads that I have been running down in my own research. However, the main concern would be whether non-myeloid cells actually upregulate IRG1 to a sufficient extent to cause the level of TCA cycle dysregulation you hypothesize.

Tomas noted that other findings from his lab showed itaconate to be a weak SDH inhibitor, which may not be expected to have a strong physiological effect at all in neighboring cells (I will not speak for him further than this, as he has also mentioned follow-up experiments that would confirm or deny this point). This may not be a problem for e.g. macrophages, where the sheer level of itaconate produced is high enough to be locally bactericidal, and therefore are more likely to be able to affect neighboring cells than any other celltype. The question then remains as to whether non-myeloid cells are able to produce enough itaconate to sufficiently alter their own metabolism, let alone their neighbors. And even if they do, the follow-up question is whether a mild auto-inhibition in 10% of cells would be sufficient to produce the level of dysfunction seen in ME/CFS. Looking through the literature, so far it seems that only cancer cells have been shown to reach a high capacity of itaconate production. Though, as you note, we might just not have looked in the right places yet.

I am currently in the process of meeting with collaborators to discuss how my own theories might be addressed through tissue samples. If I am successful in arranging a study, it may also address parts of what you propose.

Certainly, we need evidence of ACOD1 expression in sick cells, but itaconate inhibition of SDH is not the hypothesized mechanism of TCA cycle dysfunction. What matters is not the local concentration of itaconate, but rather the fraction of TCA cycle flux that is diverted to the shunt. If itaconate is rapidly converted to itaconyl-CoA and then to citramalyl-CoA and then to pyruvate + acetyl-CoA, there is no reason to expect accumulation of itaconate, but every turn of the itaconate cycle oxidizes carbon and produces at most one NADH. Indeed, if the pyruvate re-enters the TCA cycle via PC, the itaconate cycle produces zero NADH, zero FADH2, 1 CO2, and consumes 1 ATP. This is vastly inefficient compared to the normal TCA cycle, which produces 3 NADH, 1 FADH2, and 1 ATP per turn. This dramatic inefficiency is the hypothesized cause of cell dysfunction in ME.

 

[Utsikt]

This dramatic inefficiency is the hypothesized cause of cell dysfunction in ME.

I might be completely ignorant here - but what cell dysfunction are we talking about? Is it related to a specific finding or is it a general hypothesis? And would this be a downstream effect of whatever is maintaining the illness?

 

[RDP]

Very fair points, Robert @RDP.

Some rogue cell population is hiding in plain sight here I think. We are just not used to the way it is tricking everything else.

Agreed. Could be as many rogue cell populations as there were infected cells during the "trigger" infection.

 

[jnmaciuch]

Certainly, we need evidence of ACOD1 expression in sick cells, but itaconate inhibition of SDH is not the hypothesized mechanism of TCA cycle dysfunction. What matters is not the local concentration of itaconate, but rather the fraction of TCA cycle flux that is diverted to the shunt. If itaconate is rapidly converted to itaconyl-CoA and then to citramalyl-CoA and then to pyruvate + acetyl-CoA, there is no reason to expect accumulation of itaconate, but every turn of the itaconate cycle oxidizes carbon and produces at most one NADH. Indeed, if the pyruvate re-enters the TCA cycle via PC, the itaconate cycle produces zero NADH, zero FADH2, 1 CO2, and consumes 1 ATP. This is vastly inefficient compared to the normal TCA cycle, which produces 3 NADH, 1 FADH2, and 1 ATP per turn. This dramatic inefficiency is the hypothesized cause of cell dysfunction in ME.

Thank you, yes I fully understand that, my point was mainly to address the issue of “infected cells and neighboring cells” per your previous post. All the evidence I’ve seen of viral infection leading to altered metabolic environment in nearby cells is mediated through locally activated immune cells responding to the initial signal from non-immune cells (which is not itaconate, or by detecting virus on their own), as the parenchymal response would be too weak on its own. Perhaps you have seen evidence that I haven’t seen which contradicts this.

[Edit: so while I fully understand the idea of a cell-autonomous TCA cycle problem, either those 10% dysfunctional cells alone are sufficient to create the dramatic reduction seen in ME/CFS, or another explanation is needed to explain effect on neighboring cells which effectively propagates their dysfunction]

And, to extend this, the theory needs to account for a way for this hypothesized mechanism to turn from a local signal to a global one in PEM, which is the puzzle that I’ve been putting together.

 

[RDP]

I might be completely ignorant here - but what cell dysfunction are we talking about? Is it related to a specific finding or is it a general hypothesis? And would this be a downstream effect of whatever is maintaining the illness?

Theory only at this point. Theory says parenchymal cells (afferent or efferent neurons, GI SMCs, any cell type generating symptoms) are burning carbon in the itaconate cycle and therefore are producing insufficient reducing equivalents (and ATP) to carry out their normal physiological work.

During the triggering infection, the shunt is activated via PRRs. Theory says it fails to turn off. There are several potential chronicity mechanisms, but Tom's paper, which is the source of this thread, identifies one possibility.

 

[RDP]

Thank you, yes I fully understand that, my point was mainly to address the issue of “infected cells and neighboring cells” per your previous post. All the evidence I’ve seen of viral infection leading to altered metabolic environment in nearby cells is mediated through locally activated immune cells responding to the initial signal from non-immune cells (which is not itaconate, or detecting virus on their own), as the parenchymal response would be too weak on its own. Perhaps you have seen evidence that I haven’t seen which contradicts this.

[Edit: so while I fully understand the idea of a cell-autonomous TCA cycle problem, either those 10% dysfunctional cells alone are sufficient to create the dramatic reduction seen in ME/CFS, or another explanation is needed to explain effect on neighboring cells which effectively propagates their dysfunction]

And, to extend this, the theory needs to account for a way for this hypothesized mechanism to turn from a local signal to a global one in PEM, which is the puzzle that I’ve been putting together.

Among those who think every cell is an (innate) immune cell, the signal from infected cells to neighboring cells is not metabolic, but rather type I IFNs secreted by the infected cell and activating IFNARs on neighboring cells. ACOD1 is the name given to this gene after its function was identified. Its original name was IRG1 (Immune Responsive Gene 1). ACOD1 is almost surely an ISG. This is all textbook stuff. See Abbas, for example.

I totally agree any correct theory of ME must account for PEM. Not there yet with this theory.

 

[jnmaciuch]

Among those who think every cell is an (innate) immune cell, the signal from infected cells to neighboring cells is not metabolic, but rather type I IFNs secreted by the infected cell and activating IFNARs on neighboring cells. ACOD1 is the name given to this gene after its function was identified. Its original name was IRG1 (Immune Responsive Gene 1). ACOD1 is almost surely an ISG. This is all textbook stuff. See Abbas, for example.

Thank you, I am quite aware of all of this, local interferon signaling is my research focus at the moment and exactly what I was referencing in my previous post. My point re: itaconate and SDH was towards local metabolic dysregulation in neighboring cells in response to viral infection [edit: and whether that has been actually attributed to cell-autonomous responses vs. the local effect of something like itaconate coming specifically from activated myeloid cells].

Thank you for taking the time to discuss your theory on this forum.

 

[jnmaciuch]

ACOD1 is almost surely an ISG

It is, but from my prior work in macrophages I have never seen it being stimulated by IFNAR or IFNGR activation alone. In macrophages, this is induced via co-stimulation of interferons and LPS, or another TLR activator [edit: leading to both Nf-kB and STAT TF binding to facilitate ACOD1 transcription]. You would likely need to induce a strong Nf-kB response alongside IFN, which would be confounded by the normal regulators of Nf-kB in non-immune cells. Hence my skepticism as to whether parenchymal cells can upregulate ACOD1 enough to cause this metabolic dysfunction. Of course, there might be new information in the future that challenges this.

 

[Snow Leopard]

I've read this entire thread and I still don't understand how any of this is relevant to the physiology/symptoms of ME/CFS? Maybe @jnmaciuch can enlighten me?

 

[Jonathan Edwards]

I've read this entire thread and I still don't understand how any of this is relevant to the physiology/symptoms of ME/CFS? Maybe @jnmaciuch can enlighten me?

Have you watched Robert's two videos, because they are very clear. He suggests that per is due to ammonia accumulation in brain. The general theory would also predict reduced metabolic capacity for various tissues (if we think that is what we are looking for).

 

[Snow Leopard]

Have you watched Robert's two videos, because they are very clear. He suggests that per is due to ammonia accumulation in brain. The general theory would also predict reduced metabolic capacity for various tissues (if we think that is what we are looking for).

We'd have additional signs if there was ammonia accumulation in the brain and that evidence is lacking.

Ammonia easily crosses the blood-brain barrier so there would be higher levels in circulation as well.
https://my.clevelandclinic.org/health/diseases/24065-hyperammonemia

 

[Jonathan Edwards]

We'd have additional signs if there was ammonia accumulation in the brain and that evidence is lacking.

I agree, although we may not have a comparable clinical situation of pure cerebral ammonia toxicity in the context of reduced metabolic reserve.

 

[jnmaciuch]

I've read this entire thread and I still don't understand how any of this is relevant to the physiology/symptoms of ME/CFS? Maybe @jnmaciuch can enlighten me?

Agreed with Jonathan, though I think there are still additional missing pieces/contradictions with literature (which, to be clear, is to be expected from any theory so I am not disregarding it out of hand on those reasons). If this theory is correct, you would assume it leads to accumulation of the metabolites which trigger the metabosensing afferents you have brought up mediating central fatigue. As already mentioned, though, I think the link to PEM is a big missing piece here. Like you, I am not convinced brain ammonia levels are sufficient to explain for several reasons (personal opinion, take it or leave it).

I am also particularly thinking of Cara Tomas’s group and their muscle metabolism work, which showed comparable functioning between healthy and ME/CFS via galactose-only oxidative phosphorylation and fatty acid metabolism, both of which would be reliant on acetyl-CoA making it through the TCA cycle. I suppose if it’s a “weak” itaconate shunt it might be compensated by glutamine to some extent [or simply increased throughput through either pathway to counteract lack of efficiency. FAO also does produce some NADH independent of Acetyl-CoA output].

And like I said earlier, I have doubts as to whether a solely parenchymal response would be able to upregulate IRG1 to a sufficient extent. If the initial parenchymal response led to tissue resident macrophage activation during exertion, then I could see it being sufficiently propagated through those macrophages, and the local metabolic depression would be explained by itaconate inhibition of SDH alone or perhaps by those macrophages simply outcompeting parenchymal cells for glucose.

But there are doubts as to whether even high levels of itaconate from macrophages could do this in other cells, and we don’t see other signs of macrophage activation that you would expect from Nf-kB upregulation.