I believe Dr Phair is going to need to measure the amount (or relative amounts) of IDO1 protein (enzyme) in the patients’ cells versus controls. Based on the known immunomodulatory abilities of IDO2, and this includes with respect to IDO1, I think the chances are better now that the reduced consumption of tryptophan (trp) and lower production of kynurenine (kyn) he found in the small in vitro tracer experiment is due to a
decreased amount of IDO1 rather than an inhibited IDO1, which the metabolic trap hypothesis requires.
(It wouldn't be sufficient to measure mRNA in this case because the IDO1 enzyme can be subject to increased degradation under conditions related to the immunomodulatory ability of IDO2. See later below.)
For example, in this recent study:
Lipopolysaccharide shock reveals the immune function of indoleamine 2,3-dioxygenase 2 through the regulation of IL-6/stat3 signalling
it was found that IDO2 appears to act in a way that decreases the production of inflammatory cytokines, in particular IL-6, at least in mice exposed to LPS. Thus, in mice lacking the IDO2 gene, levels of IL-6 increased to significantly higher levels upon exposure to LPS than mice with a functional IDO2. (Human results relative to IDO1 and IDO2 have so far appeared largely consistent with mouse studies, I believe,)
And IL-6 has been found to drive regulatory proteolysis (i.e., degradation) of IDO1. See in particular the relevance of this study:
Deficiency of immunoregulatory indoleamine 2,3-dioxygenase 1 in juvenile diabetes
where it is stated:
IDO1 functioning is, in turn, modulated at multiple levels (i.e., transcriptional and/or posttranslational levels; refs. 11, 12, 15). In fact, the expression of IDO1 protein may not be sufficient per se for full enzymatic activity (16, 17) because IDO1 proteasomal degradation occurs in a microenvironment dominated by IL-6 (18, 19).
(My emphasis.)
The above study is very interesting because it's one of the few examples of another disease where a low intracellular (PBMCs) kyn/trp ratio has also been found, although it's also low in sera in that case as well and there are other important differences.
So, based on a (hypothetical) IL-6 related immunomodulatory scenario in which lack of a functional IDO2 results in increased local IL-6 concentrations which in turn cause an increased degradation of IDO1, it might be possible to ameliorate the effects of a non-functional IDO2 gene in some ME/CFS patients with the use of the anti-IL-6 receptor monoclonal antibody
tocilizumab. (Tocilizumab has been mentioned a few times before on this forum in various threads.)
That drug only works in a subset of juvenile type 1 diabetes patients and, intriguingly, which subset it works in may be largely dependent on a single IDO1 SNP (
rs7820268), and this may or may not be applicable in ME/CFS.
(I'm guessing that if tocilizumab were in fact useful in ME/CFS that it might not appear effective in Ron Davis’ nanoneedle unless the cells were first cultured with that drug for some period of time.)
Finally, it's obvious that even before measuring IDO1 protein levels the very first step must be to verify that the controls used for the kyn/trp tracer experiment are appropriately matched in terms of number of non-functional IDO2 alleles, i.e. their levels of homozygosity and heterozygosity for the inactivating SNPs.
The bad news is that the immunomodulatory scenario involving IDO2 makes it more likely that if appropriately matched controls weren't used then the results of the tracer experiment may actually be perfectly
normal for individuals with an inactivation of IDO2, whether they have ME/CFS or not.
The good news is that the IDO2 SNPs that yield a non-functional enzyme are quite common in the general (European ethnicity) population. Looking at the genotype data, just for the two most common ones alone, R248W (
rs10109853) and Y359stop (
rs4503083), true homozygotes are around 35% of the population, and when you add in likely effective homozygotes (i.e., individuals who are compound heterozygous for the two SNPs) together with all remaining SNP possibilities the total number of actual and effective homozygotes who have a non-functional IDO2 may be over 45% of the general population.(!)
These SNPs are thus so common that Phair may have used matched controls by sheer chance alone. (Note that it may be necessary to control for IDO1 as well, because one relatively common SNP, the aforementioned rs7820268, may afford a significant upregulation of that enzyme.)
One last little side note, I sure hope that when Dr Phair indicated that an average of 1.7 damaging IDO2 mutations per patient were found in the severe ME/CFS patient study that he meant 1.7 damaged
alleles per patient, and not actually 1.7 different IDO2 SNPs, because the latter would only matter if the SNPs were on different alleles of course.