Deep Sequencing of BCR Heavy Chain Repertoires in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome, 2025, Ryback et al

[Jonathan Edwards]

So is there like a library of antibody signatures for different pathogens and self-antigens, and the antibodies here didn't match up with anything in the library?

No, it would be hard to do that systematically.
The evidence against an antigen driven response comes I think from indicators like restricted clonality and hypermutation. When you are making antibodies to a specific antigen a higher proportion of B cells are from a few expanding clones rather than just any old random memory cells. Moreover, their antibodies show lots of secondary mutations that occur during the clonal expansion process - fine tuning if you like. My understanding is that Audrey did not find these.

Which in a way makes the VH3-30 weighting more interesting because if there is no specific antigen-driven response going on the B cell population should be evenly spread.

(I use the old VH nomenclature. It looks as if people are now writing HV. Not sure why.)

 

[wigglethemouse]

This study is listed on MapME/CFS but it doesn't include the raw data from what I can see, only summary statistics of their analyses and modelling.

The acknowledgement section of the paper states the data is available.

The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository [94] with the dataset identifier PXD016622

Here is the link - click on (more) to see the links to the actual data.
https://www.ebi.ac.uk/pride/archive?keyword=PXD016622

 

[wigglethemouse]

In case it is helpful to someone there seems to be quite a few COVID studies that have highlighted IGHV3–30 in spike neutralising antibodies. Here are the first two I came across.

A large-scale systematic survey of SARS-CoV-2 antibodies reveals recurring molecular features

Frequent use of IGHV3-30-3 in SARS-CoV-2 neutralizing antibody responses

 

[Nightsong]

Looking for any reference at all to VH3-30 (or its synonyms) in the Long COVID literature, I came across this curiosity from the recent Annesley paper:

Several Immunogloblin Light and Heavy Chain Variable (IGLV or IGHV) region genes were also present in the downregulated genes. All immunogloblin genes are located in chromosome fourteen, and as a group constituted the most frequently downregulated class of proteins in the Long COVID PBMCs. Seven out of the sixty-six downregulated genes belong to this class, namely IGLV8-61, IGHV1-18, IGHV3-48, IGHV2-5, GHV4-59, IGHV3-7, and IGHV3-30.

 

[Jonathan Edwards]

Looking for any reference at all to VH3-30 (or its synonyms) in the Long COVID literature, I came across this curiosity from the recent Annesley paper:

That's interesting but I am not at all sure what 'down regulation' of a VH gene in PBMCs would mean. A small proportion of PBMC will be B cells. The others should not be using VH genes. Only a tiny proportion of the B cells should be using it, having spliced it in during Ig gene recombination in bone marrow. The extent to which they use it will depend on their maturation status, not their commitment to the gene being spliced in. They will mostly be fresh naive cells from marrow that have not yet been through any peripheral selection and certainly not secreting Ig.

It is all a bit complicated and I can't make much of it.

 

[SNT Gatchaman]

Further to Wigglethemouse's comment 3-up, another Covid paper, looking at acute severe disease is Longitudinal Multi-omics Analyses Identify Responses of Megakaryocytes, Erythroid Cells, and Plasmablasts as Hallmarks of Severe COVID-19 (2020, Immunity) which reported —

We next analyzed the BCR repertoire in the different B cell compartments by using heavy-chain bulk- and scBCR-seq. […] Analysis of immunoglobulin heavy-chain variable region (IGHV) family subunits showed a preponderance of specific V regions in COVID-19 patients compared with in controls, e.g., IGHV3-30 and IGHV3-23 were overrepresented in PBs and neutrophil-like cells during disease.

 

[ME/CFS Skeptic]

Here is the link - click on (more) to see the links to the actual data.
https://www.ebi.ac.uk/pride/archive?keyword=PXD016622

Thanks. The data is in .raw filetype and have some trouble converting it to something like CSV or Excel. Anyone who has more experience in this?

 

[Kitty]

I can import it into both Excel (though I get a warning there's too much data to fit on one sheet) and TextEdit.

But it's not set up in such a way as to be useful. It's presumably a database, so all you get are tens of thousands of lines of not-very-useful!

 

[Hoopoe]

Thanks. The data is in .raw filetype and have some trouble converting it to something like CSV or Excel. Anyone who has more experience in this?

I haven't looked at the files, but my guess is that the .raw files are the reads of the mass spectrometer. In the Data Processing Protocol section they describe how to analyze it. There's also a download of data in JSON format which might be a more processed form of the data.

What Kitty has posted appears to be in the JSON format.

 

[Jonathan Edwards]

I have not been involved directly in VH usage weighting studies, although Jo C has, but it is looking to me as if VH30-3 may turn out to be one of these odd VH genes that has skewed rules. It might for instance always be over-represented when some cytokine is around that normally means 'I have recently seen a virus' but may give that signal falsely.

 

[EndME]

To ask the standard questions:

• Can such differences be driven by inactivity?
• Can such differences be driven by something not to do with ME/CFS (for instance more recent exposure to infection or vaccination)(I think @Jonathan Edwards has hinted at that this being a possibility, but the authors do state that they find no evidence for such an event)?
• Can such differences be driven by different storage or handling of samples or some other nuances occuring in the experimental setup?
• In what other conditions does one see such results?
• How should we follow-up this kind of research? What kind of follow-up questions such be asked and should we push for independent replication in a larger cohort?

 

[Jonathan Edwards]

@EndME,

Inactivity - vanishingly unlikely
Something other - confounding associations are a big worry but I think more for things like insulin resistance that link to general ill health. I think relatively unlikely. If samples are taken from long-standing ME/CFS recent infection does not seem likely as an explanation.
Storage handling - I think very unlikely. Storage handling is unlikely to affect proportions of gene usage. Not doing samples blinded and in parallel is always a worry but Audrey I know to be finicky about that.
A VH3-30 skew may turn up in other situations and that would be very interesting to look at.
I would follow up with a deep dive into the VH skewing literature and what is known about the signals that drive such skewing. There are people who work on this but I don't know them.

 

[mariovitali]

My favorite topic (=liver) comes up again. cc @Chris Ponting

Villous lymphocytes > 10% were detected in 50% of patients belonging to the IGHV1-2 group, in 21% of the IGHV3-23 group, and in no patient of the IGHV3-30 group (p = 0.05). Liver involvement was present in 50% of the IGHV3-30 group, in 9% of the IGHV3-23 group, and in no patient of the IGHV1-2 group (p = 0.04). HCV-serology was positive in 50% of the IGHV3-30 group, in 7% of the IGHV3-23 group, and in 17% of the IGHV1-2 group (p = 0.04). The proportion of intermediate and high risk patients according to the SMZL score was higher in the unmutated respect to the mutated group (69% vs 32%, p = 0.05).

and from another study. Note that ACLF = Acute Liver failure.

Based on the usage frequency distribution histogram of the V gene of the BCR heavy chain and clustering heat map (Fig. 3A and B), and the results obtained from the t-test for the distribution ratio of the V gene of patients with ACLF and control subjects (Fig. 3; Table II), it was found that the IGHV3-30 gene was significantly upregulated (P<0.05), whereas the IGHV3-74 and IGHV2-70 genes were significantly downregulated (P<0.05). Furthermore, a distribution histogram of the D region usage frequency of the BCR heavy chain and the D sub-genotype of each frequency clustering heat map were generated (Fig. 4A and B). The t-test for the distribution ratio of the D gene in the six patients with ACLF and six control subjects allowed for the identification of the differential gene expression between these two groups

References

https://www.sciencedirect.com/science/article/abs/pii/S1079979609000138

https://www.spandidos-publications.com/10.3892/ijmm.2018.3946


EDIT : Going through the study, we read about Gelsolin (GSN) : "GSN is a multifunctional protein that has actin binding activities. It is thought to have anti-inflammatory effects and to protect against oxidative stress by scavenging actin released from damaged cells and tissues [69]. Lower serum levels are observed in inflammatory states including acute injury [70], sepsis [71], and rheumatoid arthritis [72]. Decreased plasma levels in ME/CFS patients may reflect a pro-inflammatory environment.". Looking at reference [70] we read :

Ito H, Kambe H, Kimura Y, Nakamura H, Hayashi E, Kishimoto T, et al. Depression of plasma gelsolin level during acute liver injury. Gastroenterology. 1992;102: 1686–1692. pmid:1314752

 

[duncan]

The evidence against an antigen driven response comes I think from indicators like restricted clonality and hypermutation. When you are making antibodies to a specific antigen a higher proportion of B cells are from a few expanding clones rather than just any old random memory cells.

Does this observation hold true over long time frames involving infection, e.g. three years vs. 30 days?

 

[Utsikt]

I’m going to ask the newbie questions:

• Do we have any reason to believe that VH-30 causes any damage, or is this more of a downstream effect of something else? Both?
• If B-cell depletion doesn’t work as a treatment (neither short or long term), does that mean that the B-cells aren’t directly involved in the damage-causing part of ME/CFS?

 

[Jonathan Edwards]

I’m going to ask the newbie questions:

Excellent questions.

It is very unlikely that antibodies made using VH30-3 are specifically damaging. Indeed it is not clear that anything causes 'damage' in ME/CFS, but something is causing major suffering.
The weighting to VH30-3 doesn't tell us anything directly about what these antibodies can do. As said before it is more like which company you bought the antibody from - Nissan or Audi. The hope is that if Nissan antibodies are being used more than usual it will be a clue to what is going on, that is all.

The problem with B cell depletion as an experiment to test the role of B cells is that we already know of a number of B cell driven diseases that do not respond to B cell depletion because the antibodies involved are made by long lived plasma cells that the B cell depletion does not touch. Again, I don't think there is necessarily a damage-causing part but I suspect you are really just meaning the part causing the symptoms and disability.

 

[Utsikt]

Again, I don't think there is necessarily a damage-causing part but I suspect you are really just meaning the part causing the symptoms and disability.

I actually meant damage as in structural damage to something, but your interpretation made me realize that structural damage doesn’t have to equal symptoms and that symptoms doesn’t have to equal structural damage. Or relevant damage/symptoms.

Thank you for explaining!

 

[Jonathan Edwards]

Well spotted @Nightsong.

So, as I suspected might be the case, VH30-3 dislike VH4-34 a 'usual suspect' in this game. This certainly adds weight to the idea that these guys show up when there is some broad regulatory shift in B cell expansion.

I just can't remember what the defect is in Wiskott-Aldrich. That might tell us what the regulatory pathway is.

 

[Nightsong]

[ Originally deleted this post because I was going to add some explanation & additional references, but reposting ]

Wiskott-Aldrich - https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2014.00340/full

Castiello et al. have recently reported increased usage of VH3-30 and VH4-34 genes in WAS patients compared with controls (16), and increased usage of VH4-34, among both Cμ-expressing transitional B cells and CD21lo CD38lo B cells has been also reported (17). The VH4-34 gene encodes for self-reactive cold agglutinin antibodies (39, 40), whereas VH3-30 is highly represented among anti-platelet anti- bodies (41, 42).

Also MALT lymphoma - https://www.sciencedirect.com/science/article/abs/pii/S1044579X13001260?via=ihub
and CLL:
https://www.sciencedirect.com/science/article/abs/pii/S0037196323000999
https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0067751&type=printable

 

[Jonathan Edwards]

Which is why my post (now #38) appears psychic.

Wiskott-Aldrich is a defect on a protein (WASp) needed for forming immunological synapses, probably between T and B cells. This looks like it may be a story a bit like VH4-34. B cell clonal maturation depends on T and B cell interaction in one part of lymph node and B cell competition in another (the follicles). VH4-34 B cells don't seem to like going in to follicles and undergoing further mutation. That is normally needed for making large amounts of antibody. In lupus these cells make larger amounts of antibody, probably without needing to go in to follicles. That probably reflects the environment of complement and complement regulatory proteins.

A WASp defect might similarly shift the ease with which B cells carrying VH3-30 (I keep getting the numbers wrong but never mind) can interact in various compartments. WASp defects also lead to autoimmunity, like complement defects.

It would be so exciting if this firmed up further. Jo Cambridge and I toyed with this sort of story a few years back and concluded that maybe people with ME/CFS would have less VH4-34 usage, but that would be entirely consistent with more VH3-30 if the tweak was right.