1. Sign our petition calling on Cochrane to withdraw their review of Exercise Therapy for CFS here.
    Dismiss Notice
  2. Guest, the 'News in Brief' for the week beginning 18th March 2024 is here.
    Dismiss Notice
  3. Welcome! To read the Core Purpose and Values of our forum, click here.
    Dismiss Notice

DNA methylation and BDNF expression account for symptoms and widespread hyperalgesia in patients with CFS and Fibromyalgia, 2020, Polli et al

Discussion in 'ME/CFS research' started by Andy, Jun 21, 2020.

  1. Andy

    Andy Committee Member

    Messages:
    21,811
    Location:
    Hampshire, UK
    Paywall, https://onlinelibrary.wiley.com/doi/abs/10.1002/art.41405
    Sci hub, https://sci-hub.tw/10.1002/art.41405

    Funded by ME Research UK.

    Fukuda selection criteria.
     
    Hutan, merylg, JaneL and 12 others like this.
  2. Jonathan Edwards

    Jonathan Edwards Senior Member (Voting Rights)

    Messages:
    13,274
    Location:
    London, UK
    Mediation of symptoms by a neurotrophic factor seems and attractive idea. The only reservation I have about the result is that the difference is not very big. If a signalling molecule causes symptoms of the sort seen in ME I would tend to expect it to be twice as high or more than normal.

    The methylation difference seems to be restricted to one exon (IX). Whether or not this is expected is not clear. Perhaps it is the most relevant one for protein expression.
     
  3. Midnattsol

    Midnattsol Moderator Staff Member

    Messages:
    3,574
    But they're not saying BDNF causes the symptoms, only that it can predict them? Would you expect the same fold change in that case? A threshold of two is as arbitrary as 0.05 is for p values.

    Methylation of exons is interesting, I think in my genetics course it was brought up in regards to alternative splicing.
     
    merylg, Michelle, Ravn and 5 others like this.
  4. Jonathan Edwards

    Jonathan Edwards Senior Member (Voting Rights)

    Messages:
    13,274
    Location:
    London, UK
    I interpret 'account for' in the title as 'cause'. 'How do you account for the missing sherry?' 'Jack had a swig.'
    A softer interpretation in the sense of correlate enough to make a prediction is possible but they do talk about neurotropic factors as if they think they may be causal.

    The factor of two is not arbitrary. It is based on experience of levels of signalling mediators in disease - thyroxine, cytokines, antibodies, growth hormone, etc. Most of the time the differential is several times normal but it may be as lo as twice. A factor that is not a signalling mediator, like sodium level or haematocrit may give disease with a narrower ratio but even with glucose significant symptoms usually relate to quite marked abnormality.

    Unfortunately I think the test and control values for protein are just given as means with standard error. It would have been better to see a scatter plot.
     
  5. Marco

    Marco Senior Member (Voting Rights)

    Messages:
    277
  6. Grigor

    Grigor Senior Member (Voting Rights)

    Messages:
    540
    Is this relevant to those without FM?
     
  7. spinoza577

    spinoza577 Senior Member (Voting Rights)

    Messages:
    455
    I think the finding may be interpreted as to that the plasticity in the brain (or some brain regions) is elevated.

    This may in any sense match up with Naviaux´s finding of diminished sphingolipids, probably in some regions the actions are shut down, and in other regions highly elevated, unfortunately.

    I personally would like to prefer to have, say, a geometrical trap, with downstream effects.
    The other interpretation is of course that on any cellular level something is wrong.
     
  8. Midnattsol

    Midnattsol Moderator Staff Member

    Messages:
    3,574
    I only read the abstract and some parts of it so I might have missed the causal part.

    Let me rephrase, in -omics I've repeatedly been told the fold change is arbitrary, >2 is nice but 1.5 might also be used as a threshold. Below that you're left with the "significantly different but hardly changed"/"high fold change but not significantly different" features, like this one. But I guess there's a difference between a large proteomics dataset and looking at one protein that you know. I don't know anything about many of the metabolites or genes in the datasets I've worked with, which does not exactly help interpretation.

    I'm not sure about the glucose example, even if there are no "significant symptoms" there can be things going on at a cellular level that can be problematic long term.

    I agree with the data points, why can't everyone just supply more than the mean pr medians..

    If i remember correctly, the response to BDNF is an inverted U shape, and high levels are not necessarily good for plasticity. I have no recollection of what "high" means in this context.
     
    merylg, Michelle, Ravn and 5 others like this.
  9. Ravn

    Ravn Senior Member (Voting Rights)

    Messages:
    2,044
    Location:
    Aotearoa New Zealand
    That question occured to me, too.
    I don't understand their reasoning here. Surely if "CFS is a very heterogeneous condition" it makes no sense at all to add a second confounding condition to the mix. Especially when you're looking at BDNF, a factor implicated in the key symptom of pain in that second condition, FM. Whereas not everybody with severe ME/CFS even has significant pain.

    Also, they used Fukuda and there's little sign of PEM in their cohort. Severe ME patients would be suffering from PEM after the first assessment (travel to clinic, talking to staff, undergoing a battery of tests, etc.) so at the second assessment 3 days later you'd expect them to score much worse on the CFS symptom list (table 1) - but they don't.

    Then there's the paper @Marco linked which showed decreased levels of BDNF in ME whereas the FM/CFS cohort in the present study showed increased levels and cites other studies that have found increased levels in FM.

    All of which makes me suspect the majority of the participants were actually straight FM patients who happened to also meet Fukuda but without PEM. Speculation of course because they don't say. You'd think with all the talk about PEM in the paper they would have looked at whether their participants actually had it - and reported on it.

    The whole thing seems like a missed opportunity. The timing of BDNF is intriguing. In healthy people
    So looking at if something goes wrong with it in FM and in ME - separately! - makes sense. But they didn't include an exercise challenge (unless you count the day 1 assessment as such).
    I wonder, do they consider post-exertional pain in FM to be the same as PEM in ME?
     
  10. Jonathan Edwards

    Jonathan Edwards Senior Member (Voting Rights)

    Messages:
    13,274
    Location:
    London, UK
    If that shows the opposite maybe this is a bit of a damp squib.


    I think that relates to whether or not you think there is a real biological difference in the context of measuring lots of things scattershot. I was thinking in terms of signalling molecules actually mediating symptoms - whether directly or indirectly. In that situation I think it is pretty unusual for anything much less than a twofold difference to cause symptoms - across the board of diseases.

    I agree that glucose is arguable - but I think I did say 'even glucose' - which is not what I would consider a signalling mediator like a hormone or cytokine but rather something that mediates effects directly. I guess the logic is that signalling molecules are used as trace chemicals whose effects can be amplified by cell responses to receptors and that as such they have a several fold range of operation.
     
    Ravn, TrixieStix, merylg and 8 others like this.
  11. wigglethemouse

    wigglethemouse Senior Member (Voting Rights)

    Messages:
    954
    I tried to look at both papers, search the internet for papers on BDNF, and write a post yesterday comparing the two papers, but I'm in a bad crash and gave up.

    Anyway, with my zero biomedical knowledge I read that plasma BDNF protein measurements and serum measurements give different results. Serum. which was used for the paper of this thread title seems more stable over the day and time. It seems plasma in men has a diurnal characteristic similar to cortisol, and in women depends on menstrual age and menstrual phase, HRT etc, or something like that. The study of this thread used serum and women only.
    The study posted by @Marco used plasma. In addition they extracted PMBC's and did something with them to get a measurement of BDNF.

    So the two papers seem to have completely different methods of BDNF protein measurement. At that point I gave up.
     
    Ravn, Marco, merylg and 1 other person like this.
  12. Jonathan Edwards

    Jonathan Edwards Senior Member (Voting Rights)

    Messages:
    13,274
    Location:
    London, UK
    The actual BDNF levels in serum and plasma must be the same because serum is just plasma that has been cleared of fibrinogen by clotting. Diurnal and sex variation will apply to both. What may be different is reliability of the assays though. The only caveat to that is if BDNF can be released from white cells or platelets during clotting, in which case serum is not a good thing to use.
     
    Ravn, TrixieStix, merylg and 3 others like this.
  13. Mithriel

    Mithriel Senior Member (Voting Rights)

    Messages:
    2,815
    ME was a disease of abnormal reaction to exertion and pain, often severe. When they reinvented it as a disease of fatigue the pain aspect was seriously downplayed.

    It was only after the invention of CFS that the association with fibromyalgia was mentioned usually in the US where they never considered PEM or neurological symptoms just fatigue. Nowadays when CFS is used as a synonym for chronic fatigue and FM for widespread pain and no one a patient is likely to consult understands ME, I suspect that anyone with the pain of ME will be told they have FM as well.

    Not saying it is impossible to have FM as well as ME just that I do not see any evidence that the medical profession is taught how to distinguish the widespread agony of ME from the widespread agony of FM.

    Since the way researchers look at PEM is so bad it is difficult to know exactly who they are testing.
     
    Ravn, Amw66, oldtimer and 2 others like this.

Share This Page