Suppressed immune and metabolic responses to intestinal damage-associated microbial translocation in [ME/CFS], 2023, Uhde, Vernon et al.

Good initial sample size: 131 ME/CFS (Fukuda +CCC+ other requirements such as acute onset), 86 healthy controls

Smaller sample size for the exercise challenge: 9 ME/CFS (Fukuda only); 7 healthy controls
Plasma samples immediately before and 15 minutes after a VO2max exercise challenge
(A bit strange that the ME/CFS sample undergoing the exercise challenge is not reported as definitely having PEM as a symptom. Edit to add also, 15 minutes after the exercise challenge seems a bit short. Leaving it a bit longer might possibly have maximised differences. )

Biospecimens were frozen before analysis

 

A bit more on the sample selection
I'm not sure how well matched the samples were. They don't say they attempted to match. Reported mean ages are similar, but there is considerable variation. The percentage of females differs quite a lot, especially in the exercise cohorts where perhaps it mattered most:
Large samples: ME/CFS 68% female; HC 79% female
Exercise challenge samples: ME/CFS 89% female (8/9); HC 57% female (4/7)



Plasma assays of the large sample:

*****
IgG, IgA, and IgM anti-endotoxin (LPS)-core antibodies (EndoCAb)
IgG, IgA, and IgM antibodies to bacterial flagellin
Significantly higher in ME/CFS (large sample) on average, although certainly not for all individuals.
(The IgG is the one remembering a past infection (my mnemonic Gone); the IgM is the response to a current infection (mnemonic Momentary); the IgA is found in mucous membranes and in the gut (mnemonic Alimentary).)




*****
lipopolysaccharide (LPS)-binding protein (LBP)
soluble CD14 (sCD14)
Statistically, no difference between ME/CFS and HC. The authors suggest there is quite a difference between the humoral response (the ones above) and the acute-phase responses (these ones). I think, apart from for about a handful of outlier individuals, the differences are not so marked.

*****
Levels of IgG, IgA, and IgM antibodies to the gliadin protein fraction of dietary wheat
IgG, IgA, and IgM antibodies to casein (milk protein)
More in the ME/CFS on average, but not for all individuals. Again, some outliers are doing the heavy lifting.



*****
intestinal fatty acid-binding protein (FABP2)
Same sort of 'yeah, but meh' result for this marker of intestinal epithelial cell damage and turnover, with a small number of outliers doing the heavy lifting.

 

I think this imbalance is inexcusable now that we know from several studies that some biomedical findings differ significantly between male and female subjects.

 

Exercise sample
So, unfortunately, anything found in this sample is going to be discounted by the very small sample size. But these researchers do win points from me for using the same y axis within each pair of charts. The healthy controls charts are on the left, the ME/CFS charts are on the right. The bars join the results before and immediately after exercise.

*****
LBP and sCD14 (acute phase responses)
LBP and sCD14 increased significantly after exercise in the healthy controls; it didn't change or increased only slightly in the ME/CFS people.




*****
LPS (bacterial cell walls, they are endotoxins, that can activate immune cells, initiate clotting)
This is interesting. LPS levels in plasma unfortunately were not measured in the large sample I think. In the small exercise sample, levels of LPS in the blood were similar at rest. They did not increase significantly in the healthy controls immediately after exercise, perhaps due to the increase in LPS binding protein (LBP), which I assume is mopping up the LPS. In ME/CFS, the levels of LPS increased. For most ME/CFS individuals, the increased levels were still within the range of the healthy controls, so it wasn't a really extreme increase though. I really would have liked to have seen this measure repeated a number of times after the exercise (e.g. at 1 hour, 2 hours...), and especially repeated after another round of exercise.



*****
IgM antibodies

So, they are suggesting that, in ME/CFS, the acute phase responses are missing in action, but the humoral immune response is working hard, with increased levels of IgM to LPS, but also to dietary antigens.

Chart d is for the anti-LPS antibody; chart e is for the wheat protein antibody.



This seems to hang together in a coherent story, although no result is reported for milk protein antibodies or to flagellin. I would like to know if they tested those and, if so, what happened with them.

 

I'm not yet convinced that all of the sex differences really are as significant as some researchers suggest. I think sometimes the differences are cherry-picked sampling artefacts. But yes, the sex imbalance in the sample here is odd given the researchers aimed to look at humoral immunity.

For example:
The evolution of greater humoral immunity in females than males: implications for vaccine efficacy
2018 paper

It is possible that what we are seeing here is mostly the result of sex differences in immune responses.

 

A few more measures from the exercise group

*****
IL-10

Again, HC on the left, ME/CFs on the right. Bars join pre and post exercise results. IL-10 significantly increases in ME/CFS but not in HC.

IFN-γ and TNF-α were also looked at, but it is reported that there were no significant changes in response to exertion. It still would have been good to see the measured levels.



Metabolomics in the exercise group
I can't help but feel that the sample is way too small for meaningful results.

Figure 6c is a PCA analysis of the metabolomics of the HC and ME/CFS individuals, before and after exercise. Yes, there are differences, but it looks as though, if you took out a couple of outliers from each of the HC and ME/CFS groups, they would be looking very similar. Unfortunately.

*****
Glucose and citrate


Yes, levels of both glucose and citrate significantly increase in the plasma in response to exercise in the HC, and don't in ME/CFS. But, in such a small sample, there could easily be confounders. Eating before the exercise could cause the increase in the controls. A diabetes association makes lots of suggestions as to how avoid the increase in blood sugar, including practicing relaxation techniques before exercising - perhaps the ME/CFS participants were in fact just more chilled! The diabetes association suggests exercising later in the day, as the adrenalin effect that releases glucose into the blood is dampened in the afternoon. I found a report that males and females have different blood sugar responses to exercise, so again, we might be looking mainly at sex differences.

Adrenalin is a catecholamine. So, a sample with a higher percentage of men (as is the case in the Healthy controls compared to the ME/CFS group) could be expected to have higher adrenalin and perhaps therefore a higher blood glucose immediately after exercise. The literature seems to be a bit variable, but it seems that females have "a more precise defense of homeostasis in the recovery period post-exercise, including the control of blood glucose concentration and fuel selection (56). Due to their greater inherent capacity for lipid oxidation during exercise, women are able to regain control over glycemia and glucose flux in recovery more quickly than men (56)."

So, yeah. I'm not sure that these findings tell us much about ME/CFS. We need this replicated in a bigger sample with much better matched controls.

 

For sure DMissa, but when you have 9 cases and 7 controls and substantial variation between and within those samples, statistical methods can't do magic.