Andy
Senior Member (Voting rights)
@DMissa , is it this research that you will be talking about at UK: Conference: MitOX 2024, 12th April 2024
In vitro B cell experiments explore the role of CD24, CD38 and energy metabolism in ME/CFS, 2023, Armstrong et al
- Thread starter EndME
- Start date
I have long been of the view that most of what we see and know so far about ME is the patient's physiological and behavioural reaction to the underlying problem, the attempts to deal with a pathological demand on otherwise normal healthy systems now being forced to operate beyond their sustainable capacity, not the underlying problem itself.
ME/CFS Science Blog
Senior Member (Voting Rights)
For anyone struggling (like me) to find the sample size:
A total of eight ME/CFS patients (six female (F) and two male (M); median age, 33 years; range, 22–52 years) and seven healthy controls (HC) (five F and two M; median age, 33 years; range, 23–63) were included.
DMissa
Senior Member (Voting Rights)
That is a hard question to answer because there is a lot going on and not sure exactly which part of it you are referring to (whether my comment or the paper). I will do my best to clarify. I will do so only briefly so as not to derail away from this paper since it mostly concerns my own largely unrelated work, even if Chris is involved.
1) I had and have no involvement with this paper + data nor am I aware of being involved with following up on it directly. I met Jo and Fane while the work was being done but that was about it. The cell lines I have published about are selected from B cells but they are transfected with EBV and imo not directly comparable. I am working with Chris on these lines but I would consider the work separate for this reason. When I said where "we" are at I meant the community in general, from my POV anyway.
2) If you refer to me mentioning cause-effect and mechanism determination, I may have given the wrong impression that it was directly tied to this paper. Maybe not, but I will clarify in case. I was referring to work that I am involved with, while assuming that others are maybe doing similar things (following up published differences by testing how they relate to one another and to pathology: in isolation, in combination, by eliciting them by different approaches, and by using several types of cells, tissues, and even organisms). I don't know if there is such work related to this paper being undertaken. Basically again my comments were general (about this area of the field).
3) The MitOX talk will likely include data collected in collab with Chris. Some may be published by then, first manuscript almost finished. Depending on what makes the cut, aspects of the presented work may be relatable to this paper (since we are also working with unimmortalised immune cells too, as well as other tissues). Trying as hard as I can to get everything written up so regardless it should be in everybody's hands at some point.
PS: I do hope that the good work in this paper is being followed up. Jonathan might know? It may also be private information. Anyway I say this because as a generalist cell biologist I want more people who actually know what they're talking about looking at the metabolism of immune cells in such close detail.
Hopefully this is all clear now.
DMissa
Senior Member (Voting Rights)
Happy to report that upon another look this paper seems to replicate closely with my own work
I just checked the unadjusted* mito red fluorescence of about 100 ME/CFS LCLs and something like 60 or 70 HC and it is also reduced as seen in the B cells in this paper. Sharing it here because that value probably won't get reported in a paper or used for anything by itself.**
In fact if we take the three things directly related to energy pathways from this paper: 1) mito red fluorescence is down 2) amino acid usage is favoured 3) glycolysis unaffected, all three of those things match the published cell line data. Could suggest that these specific differences persist thru into the lines despite the epigenetic changes induced during immortalisation. Would have been good to check this directly in a fraction of my parental B cells... hard to directly compare as I said in the last comment. But a good sign.
*Mito red was used by itself in this study which means the fluorescence is a function of both mass + MMP.
**Because MitoTracker Red stains membrane but is also MMP dependent (when presenting mass I use mito green, and presenting MMP I use mito red normalised to green to produce a MMP measurement that accounts for differences in mass)
I just checked the unadjusted* mito red fluorescence of about 100 ME/CFS LCLs and something like 60 or 70 HC and it is also reduced as seen in the B cells in this paper. Sharing it here because that value probably won't get reported in a paper or used for anything by itself.**
In fact if we take the three things directly related to energy pathways from this paper: 1) mito red fluorescence is down 2) amino acid usage is favoured 3) glycolysis unaffected, all three of those things match the published cell line data. Could suggest that these specific differences persist thru into the lines despite the epigenetic changes induced during immortalisation. Would have been good to check this directly in a fraction of my parental B cells... hard to directly compare as I said in the last comment. But a good sign.
*Mito red was used by itself in this study which means the fluorescence is a function of both mass + MMP.
**Because MitoTracker Red stains membrane but is also MMP dependent (when presenting mass I use mito green, and presenting MMP I use mito red normalised to green to produce a MMP measurement that accounts for differences in mass)
I have a couple questions for you regarding the B cell work..
- CD38 is known to degrade NAD+, removing the NAD+ pool would reduce the speed of ATP production, do you think Daratumumab may have potential by simply removing cells that are more likely to reduce the NAD+ pool?
- CD24 seems to have been identified consistently elevated on B cells in ME cohorts by Jo Cambirdge, no one else is yet to even look at it, what is your opinion on why that could be?
Jonathan Edwards
Senior Member (Voting Rights)
I don't really see the connection. Like CD20, CD38 is just a useful ligand for identifying cells for antibody-mediated killing. If you kill lots of plasma cells you end up with no antibody production but not much effect on the rest of the body as far as I know.
I see that CD38 seems to be a significant NADase, but I am unclear what the significance of that is to metabolic function. In 2000 we didn't use anti-CD38 because we were told CD38 was expressed on other cells, such as cardiomyocytes. It seems to be that anti-CD38 antbodies do not cause major problems with these other cells. So presumably their metabolism carries on as usual.
Jo was interested in CD24 because in repopulation following rituximab CD24 return behaved strangely. I think particularly in patients with thrombotic thrombocytopenia purpura. There aren't really any other B cell biologists working on ME/CFS as far as I know, now that Amok Bansal has largely retired from research. Even most other immunologists probably have little interest in CD24.
If B cells are relevant to ME/CFS I suspect the story is subtle and not a matter of some obscure specific autoantibody effect. Something that has proven important in lupus is the control of extra follicular B cell survival. It may also be relevant in RA in a different way. The question arises as to whether in ME/CFS the female predominance reflects a shift in maturation control for B cells, analogous to but not the same as in lupus, that allows for the formation of antibody populations that are not autoreactive but 'bad' for some other generic reason. If they were directed against foreign antigens and made by long lived plasma cells then rituximab would make no difference.
Maybe there is another 'quality control' mechanism for antibodies that is defective in people with ME/CFS such that they form antibodies to microbes that get in the way in some non-specific way - maybe by forming complexes of the wrong size or triggering too much ADCC or something. These antibodies might normally be selected against during an immune response through some interaction with CD57+ T cells outside follicles or something (just an illustration).
CD24 might be a marker of such a maturation shift, which might of course also be reflected in a shift in mean values for B cell metabolic pathways.
Thank you. I agree with you on the role of B cells in the disease. I'm more interested in their unusual behaviour, which may be due to the underlying problem or a reflection of that problem.
DMissa
Senior Member (Voting Rights)
ay, there’s the rub
edit: Hi Chris, I didn't realise this was you. No wonder you were making so much sense. HAHAHA
Last edited:
ryanc97
Senior Member (Voting Rights)
Hallo @Jonathan Edwards, @DMissa,
Do you think this increased expression of CD38 on stimulated B cells could have any link to the Daratumumab trial results?
Because it looks like the % difference on CD38 is quite large. For example, in the chart below, on day 3 of culture of the B cells, whatever that means, the median HC CD38 level was 45% but the median ME CD38 level was 75%. That's a large difference.
To me this links to two things:
1. Dara is potentially depleting some B cells since they have CD38 in ME patients. But this is not important as Rituximab already depletes B cells without much effect.
2. Somehow the ME/CFS LLPCs have more CD38. Which makes them better targets for Dara. I think this is a very hopeful stretch. Any LLPCs that came from B cells, post-ME, "inherit" this higher expression of CD38.
Do you think this increased expression of CD38 on stimulated B cells could have any link to the Daratumumab trial results?
Because it looks like the % difference on CD38 is quite large. For example, in the chart below, on day 3 of culture of the B cells, whatever that means, the median HC CD38 level was 45% but the median ME CD38 level was 75%. That's a large difference.
To me this links to two things:
1. Dara is potentially depleting some B cells since they have CD38 in ME patients. But this is not important as Rituximab already depletes B cells without much effect.
2. Somehow the ME/CFS LLPCs have more CD38. Which makes them better targets for Dara. I think this is a very hopeful stretch. Any LLPCs that came from B cells, post-ME, "inherit" this higher expression of CD38.
Attachments
Jonathan Edwards
Senior Member (Voting Rights)
Yes, but it is hard to say exactly what.
Binding CD38 may have important effects on B cells other than killing.
I think the rapid increase in CD38 is because the b cells were primed in ME, primed because of chronic stress signals from cytokines or something else (perhaps even genetic)
In vivo it may be that ME create more short lived plasmablasts in response to even mild triggers for B cell activation. More short-lived plasmablasts from inappropriate activation could mean that patients also get more long-lived plasma cells. Especially if the checkpoints are also less.
I assume Dara impacts the long lived plasma cells if it's B cells or it impacts CD38 on other cell types.
@Jonathan Edwards
Could this be a reasonable pulling together of different findings:
CD38 functions as a key co-receptor in the BCR complex, associating with CD19 and modulating BCR signaling by lowering activation thresholds, enhancing calcium fluxes, and amplifying responses to antigens—including weak/self-antigens. In the primed state, rapid CD38 upregulation further sensitizes BCR signaling, promoting survival, proliferation, and differentiation of clones that might otherwise be tolerized. This favors expansion of autoreactive or skewed clonotypes (like IGHV3-30-enriched ones, common in naive repertoires but overrepresented in effectors here due to tolerance breaches), leading to the observed oligoclonality and pathogenic skewing that feeds long-lived plasma cells. This then also explains daratumumab's efficacy in depleting the downstream CD38^{high} pathogenic effectors.
Now the same thing to prime the CD38 also opens up more niches for plasma cells. This chronic inflammation could be set up by having a chronic infection phase or a drawn out acute phase. Perhaps it starts with post viral fatigue syndrome but then those that get ME/CFS are the ones that get this CD38 priming and in appropriate plasma cells in more niches. Those that don't get the priming and niches simply recover from the PVFS and never meet ME/CFS.
Could this be a reasonable pulling together of different findings:
CD38 functions as a key co-receptor in the BCR complex, associating with CD19 and modulating BCR signaling by lowering activation thresholds, enhancing calcium fluxes, and amplifying responses to antigens—including weak/self-antigens. In the primed state, rapid CD38 upregulation further sensitizes BCR signaling, promoting survival, proliferation, and differentiation of clones that might otherwise be tolerized. This favors expansion of autoreactive or skewed clonotypes (like IGHV3-30-enriched ones, common in naive repertoires but overrepresented in effectors here due to tolerance breaches), leading to the observed oligoclonality and pathogenic skewing that feeds long-lived plasma cells. This then also explains daratumumab's efficacy in depleting the downstream CD38^{high} pathogenic effectors.
Now the same thing to prime the CD38 also opens up more niches for plasma cells. This chronic inflammation could be set up by having a chronic infection phase or a drawn out acute phase. Perhaps it starts with post viral fatigue syndrome but then those that get ME/CFS are the ones that get this CD38 priming and in appropriate plasma cells in more niches. Those that don't get the priming and niches simply recover from the PVFS and never meet ME/CFS.
Jonathan Edwards
Senior Member (Voting Rights)
I think that would be a fair suggestion. It needn't involve any inflammation, just antibodies misbehaving in signalling systems. It would fit with the FcR1 idea, although I don't have any great investment in that other than as an example.
It would be nice if TLR7 or 8 could be involved, although BTK being on the X chromosome might be enough to explain increased risk in women.
One thing Jo and I talked about in relation to unwinding autoimmune loops was that it would be good to discourage ongoing re-education of new B cell clones by old B cells clones through blocking a signalling route rather than just having to kill everything as for ritux. If Dara were to dispose of long lived plasma cells producing unwanted antibody AND block ongoing re-education of B clones destined to make new unwanted antibody the 'reset' button might do its job.
NK cells might be important for plasma cell killing but not needed for the other bit.
That of course was in the context of traditional autoimmunity but we always assumed that there would be variants on the theme that did not necessarily involve classical autoantibodies - maybe including MS.
It could certainly for the FcR1 idea. It's in the same vein really.
I've recently been playing with the idea that PVFS or an elongated acute infection may actually be the trigger point for ME/CFS and that the mechanism of ME/CFS is produced by that elongated inflammation event as part of the long acute infection.
I think this is interesting because women are have a larger inflammatory response to infection on average and it's suggested that there may be a 2-3 times higher prevalence of PVFS in women. This would mean that the mechanism that switched on ME/CFS may not be female specific but females are more likely to have the trigger ongoing inflammation event. Tlr7 likely related.
I'm not saying that ME/CFS has ongoing inflammation, I'm more suggesting that PVFS does and it might be that what makes ME/CFS is priming the B cells to have more sensitive CD38, which in turn increases the pathogenic plasma cells. As you say, they may perpetuate by "training" the B cells. I guess it begs the question of what's happening with plasma cells in bone marrow.
I'm also really interested in the orexin and CRH link now too. I'm wondering if orexin impairment could also prime the CD38 on B cells.
Are there any drugs that raise NK cells that could be used prior to Dara? Like a combo treatment.
Kitty
Senior Member (Voting Rights)
Might there be answers in cancer studies, where some treatments are aimed at increasing the effectiveness of NK cells by priming them?
The apparent lack of response to dara (which might still need confirming) may not purely be down to numbers.
ETA: If dara pans out, it might be enlightening to do a small study on people found to have low counts on initial assessment. Do their low numbers persist, or is it just a timing issue?
Utsikt
Senior Member (Voting Rights)
Can you define what you mean by inflammation is this context? Is seems to mean something different to every person, so it’s difficult to keep track.
jnmaciuch
Senior Member (Voting Rights)
What is hard for me to reconcile about this proposed higher CD38 induction —> more plasma cells problem is that it would only make sense if you also had a magic mechanism like in the Fcgr1 hypothesis explaining how the antibodies themselves, once produced by these plasma cells, invoke a response that leads to ME/CFS symptoms without all the simultaneous inflammation-inducing signaling pathways that you normally get from antibody binding and immune complex stimulation.
In the thread for the hypothesis paper, @Jonathan Edwards proposed an as-of-yet-undiscovered mechanism by which immune cells bearing Fcgr1 get activated at a lower threshold of antibody binding and selectively shuttle right to the brain, releasing cytokines and causing problems there without showing telltale signs in the circulation or anywhere else prior studies have looked. Unless that undiscovered “stealth” Fcgr1 mechanism also happens to release signaling molecules that globally modulate B cells (but haven’t been detected in any prior study), this CD38 theory effectively requires 2 separate, enduring problems in ME/CFS.
I could understand if maybe there was a lingering epigenetic shift in B cells in the months after infection leading to higher CD38 induction, making someone more likely to develop the plasma cells that eventually maintain a disease state. But this paper shows that aberrant CD38 induction persists in the circulating cells of pwME who have been sick for years. So you’d need a way to explain both how CD38 mediates the disease state, and the disease state maintains higher induction of CD38. That’s why the idea of CD38 enhancing the interferon response during calcium flux (via mtDNA release) and being an interferon stimulated gene itself made sense to me more than an antibody theory.
Last edited:
Jonathan Edwards
Senior Member (Voting Rights)
Worth remembering that in many autoimmune diseases antibodies do not cause inflammation. They do other things. RA is the paradigmatic small immune complex disease, involving TNF, but lupus is mediated through a whole lot of other pathways, including effects of large complexes that do not appear to involve TNF (CRP is often normal). And in conditions like scleroderma the problem is unlikely to be immune complexes at all. Antibodies may be interacting with topoisomerase-1 actually within nuclei.
25 years ago everyone was making the mistake of thinking that immune complexes must work through fixing complement. We realised that in RA the problem is likely to be that they fail to fix complement so do not get cleared by CR1 and have the opportunity to engage FcRIII. People are remarkably slow to take on board shifts in understanding of this sort. Another shift of understanding would be entirely in order.
There is no need to invoke inflammation in antibody mediated disease - look at Ig-based amyloid nephropathy for instance.
jnmaciuch
Senior Member (Voting Rights)
Sure but in those cases we see progressive tissue damage and [edit: constant] differences in immune populations even just doing a single cell screen of circulating or local cells at the site of damage. My point is just that we have no apparent sign whatsoever of any abnormality that can be traced back to antibodies other than the feeling of being sick. [Edit: it would be a magic bullet solution indeed if there was a way that antibodies caused a problem that only lead to the feeling of being sick, by a completely different and unknown mechanism than the ones we already know about that mediate sickness behavior during infection.]
Last edited: