I've had to rethink what I wrote above, because I forgot that PPAR delta (PPARd) was found to be upregulated in the recent Fluge & Mella study.
I think the finding of increased PPARd is actually the strongest support (in terms of metabolic findings) for an argument of FXR being over-activated in ME/CFS, rather than the increased PDK4 that Fluge & Mella also found. (FXR increases PPARd which in turn increases PDK4, but PDK4 could have been increased in a number of other ways, and other PDKs were found to be increased by Fluge & Mella as well.)
But if FXR is being over-activated, then I can't currently explain why ceramides are low. Conversely, if FXR is being under-activated, I can't explain why PPARd is high.
So I guess for now it might be somewhat of a toss-up whether FXR is under- or over-activated (or neither) in ME/CFS. Hopefully
@eljefe19's experiment will be revealing one way or the other.
FWIW, there are a couple of interesting mouse studies, one of which might support under-activation, the other over-activation - assuming the same thing happens in humans.
For under-activation:
In this
study looking at the effects of the synthetic FXR antagonist, glycine-β-muricholic acid (Gly-MCA), on the mouse intestinal microbiome, that antagonist caused a reduction in the ratio of Firmicutes to Bacteroidetes bacteria. That effect was blocked when an FXR agonist was used in combination with Gly-MCA, i.e., their opposite actions cancelled each other out.
This type of reduced ratio of Firmicutes to Bacteroidetes is what has been found in ME/CFS in at least 3 different studies, I think. (See, e.g.,
Reference.)
So that mouse study may be suggesting that an FXR antagonist could take the ratio in the wrong direction in ME/CFS, and that the use of an agonist instead might increase the ratio of Firmicutes to Bacteroidetes phyla, perhaps then helping to normalize the microbiome.
And if it's true that FXR has a significant effect on ceramide levels, then agonism may fix the problem of the low ceramides found in Naviaux's study and the downstream consequences.
On the other hand, in favor of the over-activation scenario:
This
study used the FXR agonist, obeticholic acid, in a common mouse model of multiple sclerosis known as "experimental autoimmune encephalomyelitis." Obeticholic acid was found to reduce the symptoms of the disease.
And MS and ME/CFS appear to be sort of opposite diseases in major respects, with an over-activated mTORC1 pathway and decreased Tregs in MS (i.e., too much immunostimulation), and the opposite in ME/CFS (i.e., too much immunosuppression).
The study is indicating that obeticholic acid had an immunosuppressive effect helpful for the MS model and therefore possibly bad for ME/CFS, suggesting that an FXR antagonist might be the correct therapy. Unfortunately, there are no antagonists that are both specific to FXR and also very potent that have been trialed in humans. The Gly-MCA mentioned above might be a candidate, though. (Note that humans don't make the tauro-beta-muricholic acid mentioned in the original post of this thread.)
Again, though, whether either of those studies seemingly supporting over- or under-activation of FXR apply to humans is another matter.