Plasma Extracellular Vesicle Surface Marker Profiling Reveals Immune Cell–Associated Mitochondrial Membrane Potential Alterations…, 2026, Ikeda+

[SNT Gatchaman]

Plasma Extracellular Vesicle Surface Marker Profiling Reveals Immune Cell–Associated Mitochondrial Membrane Potential Alterations in Long COVID and Myalgic Encephalomyelitis/Chronic Fatigue Syndrome

Spoiler: Authors

Ikeda, Gentaro; Koike-Ieki, Mariko; Inoue, Hiroyuki; Dadhania, Arya V; El Kamari, Vanessa; Jagannathan, Prasanna; Geng, Linda N; Miglis, Mitchell G; Shafer, Robert W; Yang, Phillip C; Bonilla, Hector Fabio

BACKGROUND
Long COVID (LC) is characterized by symptoms persisting at least 3 months after SARS-CoV-2 infection and affecting multiple organ systems. Diagnosis relies on subjective criteria without established biomarkers. Immune dysregulation and mitochondrial dysfunction are implicated in LC pathophysiology. Given clinical overlap with myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS), we investigated whether plasma extracellular vesicles (EVs) capture shared molecular signatures.

METHODS
Plasma EVs from 125 individuals across pandemic-era and prepandemic cohorts were analyzed. The pandemic-era cohort included COVID-Recovered, LC with ME/CFS phenotype (LC-ME/CFS), and ME/CFS without infection (pan-ME/CFS). The prepandemic cohort included ME/CFS and matched controls. Extracellular vesicles were isolated using size-exclusion chromatography. Concentration and size were assessed by nanoparticle tracking analysis, and surface markers and mitochondrial membrane potential were evaluated by flow cytometry.

RESULTS
Both pan-ME/CFS and LC-ME/CFS exhibited elevated EV concentrations compared with COVID-recovered controls after false discovery rate (FDR) correction (q = 0.0042 and 0.0024). Leukocyte-, monocyte/macrophage-, and platelet-derived EVs were increased, whereas B cell–derived EVs were reduced in both groups. Compared with controls, pan-ME/CFS demonstrated increased mitochondrial membrane potential in B cell–, monocyte/macrophage-, and NK cell–derived subsets after FDR correction, whereas no significant differences were observed in LC-ME/CFS. Prepandemic ME/CFS showed a nominal increase in leukocyte-derived EVs that did not persist after correction, whereas elevated mitochondrial membrane potential in B cell–derived EV subsets remained significant.

CONCLUSIONS
ME/CFS and LC-ME/CFS demonstrate partially overlapping immune cell–associated EV alterations. Mitochondrial membrane potential alterations within selected immune-derived EV subsets, particularly B cell–associated EVs, suggest immune-metabolic involvement. Plasma EV profiling may inform future biomarker development.

Web | DOI | PDF | Open Forum Infectious Diseases | Open Access

 

[DMissa]

Unless I am missing something I am not sure that mitochondrial membrane potential has been measured correctly. It is a frustratingly common pitfall. Mitotracker red fluorescence is proportional to mitochondrial membrane potential, yes. But if you have more or less membrane actually present you will accordingly have more or less signal. You need to stain for total mitochondrial membrane mass to normalise the membrane potential signal back against. This does not appear to have been done, so these measurements could reflect more or less mitochondria inside the EVs by chance or some other reason.

It is like comparing the light levels in two rooms and determining that the lightbulbs in one room are more powerful because the room is more well lit, but without counting whether the number of lightbulbs is the same between the rooms.

There is the bigger picture biological question around what the relationship is between mitochondria within EVs versus inside an intact cell but the methodological concern is more pressing.

 

[Hutan]

Sadly, there may be issues with the control groups that are confounding results. For the pre-pandemic samples, it seems the age and sex of the donors is not known.

From Table 1

Post-pandemic
21 LC-ME/CFS............ mean age 48.0; 85% female
29 Pan-ME/CFS........... mean age 43.5; 75% female
25 COVID-recovered.... mean age 41.0; 24% female

Pre-pandemic
25 Pre-ME/CFS............ mean age not available; female % not available
25 Healthy controls....... mean age not available; female % not available

Edit: (although, I see further on in the paper that the entire Stanford pre-pandemic biobank's means was: 49.9 years and approximately 76%, so, with some luck, the samples could be comparable)