Apologies for taking a long while writing my email to the paper's lead author, thankfully
Xinnan Wang replied almost immediately, covering all the question points I had summarised (apologies if I failed to ask something or sounded dumb at all).
It seems that most of the answers come down to the fact that the investigating lab were working with the samples made available to them by the consortium. These came from the Biobank funded by the Chronic Fatigue Initiative set up by the Hutchins Family Foundation (previously used in the
Lipkin,Hornig 2015 cytokine study), including samples from the CFS clinics of doctors Klimas, Bateman, Peterson, Levine, Felsenstein and Komaroff (as I'd missed stated in the paper). Also, Xinnan has been in touch with Hanson (directly, I think) which is great to hear. See her full reply after the break, below (with my emailed questions indented and italic).
Thanks for your email. I did a quick read through of the blog and wow I was amazed by the responses including from both scientists and patients. I’m very glad to see our paper triggered such heated responses and I hope the study is useful for paving roads for new future directions.
A little background of this study may already answer some of your questions. This study was part of a collaborative effort or a consortium by many CFS experts and clinics and the Chronic Fatigue Initiative. The goal for my lab was to identify potential mitochondrial phenotypes in patients' samples collected by the consortium. Therefore, my lab was not involved in the sample collection and storage. The consortium could only provide frozen PBMCs from patients and controls at the time, so it was not possible to test it in other cell types. Here are my answers to your specific questions:
(1) Why was there little or no discussion about whether or not measurements of PBMN cell mitochondria would be representative of all subject's cells in general? (Was any content cut from the paper, for length, or such? The ultimate conclusion seemed somewhat unsupported.)
Although PBMCs do not necessarily represent the other cell types, they are mostly accessible and have been used in similar studies. Because our results were opposite to some studies (e.g. Castro-Marrero et al., 2013 using the same cell type), we were trying to emphasize the difference rather than attempting to conclude the causes. Our result shows that ATP levels are increased in PBMCs from CFS patients which suggests that the symptoms may not be caused by lack of ATP in PBMCs.
Yes we have also discussed this with Dr. Hanson. Please note that glycolysis shouldn’t be activated in immune cells in the control subjects because their immune system is not activated, and the potential increased glycolysis in the patients suggest that their immune cells may be activated. This is definitely a very interesting hypothesis and further studies should be conducted.
(3) Was the team aware of the potential for altering the PBMN cell's characteristics by having them in a growth medium (for 2-3 days, was it?), free of any factors that might be found in subject's serum? (The word from a number of metabolomics researchers is that they are sure there is something in patient's blood that is inhibiting mitochondria and are now trying to isolate this factor(s).)
Yes of course these cells are very different from in vivo. However, as explained in the background this was the only option at the time and even if cultured in vitro these cells are still useful for studying mitochondrial function and ultrastructure.
(4) Were your team aware of any other factors that might have selected for PBMN cells of different function in persons with ME/CFS? (Dr Jonathan Edwards talked of potential cell age differences, depending on amount of exercise dependant muscle repair taking place, for example.)
Unfortunately our part of study did not involve any clinical manifestations of these patients involved. I cced Dr. March here who may be a better person to answer these questions. We received these cells in a limited amount (because of the IRB and because these cells were distributed to multiple labs) which was insufficient for us to measure both the features and compositions of these cells and the mitochondrial parameters shown in the paper.
(5) Your paper referenced Dr Myhill et al's 2012 and 2013 papers on mitochondrial dysfunction tests (references 10 and 11). Were there any thoughts on the validity of those tests and any reason your team used PBMN cells, rather than attempt to replicate the older findings which used neutrophils?
Again, great question but we were limited by what we could get (limited number of PBMCs per subject). So when we designed the experiments, we tried to maximize the mitochondrial parameters we could measure and meanwhile generate results that would be new to the field (such as mitochondrial ultrastructure and complex activities) rather than replicate one particular study exactly which already used different cells from a different cohort.
(6) Bonus (not presented in my thread post): do you have any information you could share about the functional severity of the patients sampled and/or their illness duration? (Also, the matching of the controls, presumably age, gender, etc, but not for activity level?)
Unfortunately I was not involved in the clinical part of the study and not part of the consortium that could share this information. Dana could further discuss this question with you.
Best,
Xinnan