Mitochondrial function is impaired in long COVID patients

Abstract

Background: The Long COVID syndrome is a major global health problem, affecting approximately 10-20% of individuals infected with SARS-CoV-2 virus with many remaining symptomatic beyond one year. Fatigue, reduced exercise tolerance and hyperlactataemia on minimal exertion have led to the suggestion of a bioenergetic defect. We hypothesised that mitochondrial dysfunction is a pathological feature in Long COVID cases and would correlate with clinical outcome.

Methods: This prospective, case-controlled, observational study recruited 27 participants with an established diagnosis of Long COVID syndrome from a single tertiary clinic together with 16 age-matched controls aged 25-65 years. Seahorse-based mitochondrial flux analysis and bioenergetics profile of isolated peripheral blood mononuclear cells (PBMCs) was performed and correlated with clinical phenotype.

Findings: Long COVID cases had an increased baseline and ATP-induced oxygen consumption rate with a significant attenuation in tetramethylrhodamine methyl ester perchlorate fluorescence response to oligomycin. Correlations were observed between mitochondrial function and autonomic health, quality of life and time from index infection. Sex-specific differences were also observed.

Interpretation: PBMCs from Long COVID subjects exhibit an exceptional and distinctive change in ATP synthase, as it contributes to the mitochondrial membrane potential rather than using it exclusively to generate ATP. The findings suggest that the enzyme runs both forward and reverse reactions, synthesising and hydrolysing ATP. The correlation of mitochondrial function with clinical phenotype in Long COVID may indicate a causal relationship and warrants further validation in larger scale studies.

Keywords: ATP synthase; long COVID; mitochondrial dysfunction; peripheral blood mononuclear cells (PBMCs).

Graphical abstract

Figure 1.

Comparison of oxygen consumption rate (OCR) of PBMCs collected from long-COVID cases (n = 27) and controls (n = 16). Measurements were performed by seahorse analyser and OCR was corrected by post-run cell number determination. (A) The basal OCR; (B) shows the ATP-linked OCR. (C) and (D) illustrate proton-leaked OCR and extracellular acidification rates (ECR) between the 2 groups, respectively. (E) and (F) represent tetramethylrhodamine methyl ester perchlorate (TMRM) staining and the mitochondrial membrane potential (ψm) respectively (data is expressed as median with interquartile range, and medians are compared using the Mann–Whitney U test) A.U. Arbitrary Units.

Figure 2.

Comparison of mitochondrial biomass parameters of peripheral blood mononuclear cells from patients (n = 27) and controls (n = 16). Panels A and B, respectively, represent percentage citrate synthase activity and TOM20 protein expression (data is expressed as median with interquartile range, and medians are compared using the Mann–Whitney U test).

Figure 3.

Comparison of tetramethylrhodamine methyl ester perchlorate (TMRM) dynamic of peripheral blood mononuclear cells collected from patients (n = 27) and controls (n = 16). Panel A–C show the change in TMRM upon addition of oligomycin, FCCP and antimycin/rotenone respectively. Panel D illustrates the changes in TMRM by histograms (data is expressed as median with interquartile range, and medians are compared using the Mann–Whitney U test).