A review published in Medicine Discovery argues that mitochondrial oxidative stress sits at the centre of a self-reinforcing loop with immune activation in rheumatoid arthritis, systemic lupus erythematosus and multiple sclerosis, and that because the obstacles to treatment differ by disease, therapies aimed at a single pathway may be compensated for by others
Mitochondria do more than supply energy. In immune cells they also help decide whether a cell becomes activated, stays quiet, or turns against the body, and their by-products participate directly in the signalling that shapes an immune response. A review published in Medicine Discovery examines how one consequence of mitochondrial dysfunction — oxidative stress — contributes to three prototypical autoimmune diseases: rheumatoid arthritis (RA), systemic lupus erythematosus (SLE), and multiple sclerosis (MS). The review argues that mitochondrial oxidative stress and autoimmune activation reinforce one another in a loop, and that the barriers to treating it are not the same in each disease.
Rheumatoid arthritis, systemic lupus erythematosus and multiple sclerosis together represent a substantial global health burden. Each is driven by aberrant immune activation that produces chronic inflammation and multi-organ damage, and each has a pathogenesis that is intricate and not yet fully understood. As the authors write, "Specifically, we move beyond listing pathways to explicitly construct an integrated regulatory network, highlighting the functional cross-talk between NRF2, AMPK, SIRT1, and mitophagy." It then compares mitochondrial pathomechanisms across the three diseases and assesses how far the resulting therapeutic strategies have actually travelled.
Where the oxidative load comes from
Mitochondrial oxidative stress arises when the generation of mitochondrial reactive oxygen species (ROS) outpaces their elimination. Under normal conditions ROS act as signalling molecules in metabolism and immunity; when production exceeds the capacity for clearance, they damage proteins, lipids and DNA. The main source described in the review is leakage of electrons from complexes I and III of the mitochondrial electron transport chain, which increases the generation of superoxide anion. Abnormal activation of NADPH oxidase (NOX) and xanthine oxidase amplifies ROS production further, although the review is careful about the boundary between compartments. As the authors note, "It is important to note that while NOX family enzymes and xanthine oxidase are significant sources of cellular ROS, they are not mitochondrial in origin." Their activation can nevertheless influence mitochondrial function indirectly through redox crosstalk.
The other side of the balance is a weakened antioxidant defence system: reduced activity of superoxide dismutase, catalase and glutathione peroxidase, together with lower levels of non-enzymatic antioxidants such as glutathione and coenzyme Q10. The consequence is accumulation of ROS through inadequate clearance, followed by structural damage that includes mutations in mitochondrial DNA, peroxidation of membrane lipids, and oxidative modification of respiratory chain proteins, along with reduced mitochondrial membrane potential and impaired ATP synthesis.
Five control systems, one network
The review organises the regulatory landscape into five pathways, ordered by how they act on mitochondrial redox homeostasis. The NRF2/KEAP1-ARE pathway and the AMPK/SIRT1/PGC-1α pathway regulate antioxidant gene expression and mitochondrial biogenesis. SIRT3-mediated deacetylation acts directly on mitochondrial proteins to fine-tune electron transport chain function and antioxidant enzyme activity. Mitophagy removes damaged mitochondria before they can release ROS and mitochondrial DNA. The inflammasome pathway represents the inflammatory response that follows when these upstream defences fail. As the authors put it, "This ordering reflects a logical progression from ROS prevention and detoxification through organelle quality control to immune activation."
The review's central methodological argument is that these five systems behave as one network rather than as parallel tracks. The autophagy adaptor protein p62/SQSTM1 not only mediates the degradation of damaged mitochondria but also activates NRF2 by competing with KEAP1, coupling the removal of a ROS source to the upregulation of antioxidant defences. The NAD + pool jointly controls SIRT1 in the nucleus and SIRT3 inside mitochondria. Under energy stress, AMPK both stimulates PGC-1α directly and enhances NAD + biosynthesis through NAMPT, engaging SIRT1 and SIRT3 together. AMPK and mTORC1 act as opposing, context-dependent regulators, so that one kinase is activated when energy is scarce and the other when nutrients are abundant. The practical implication, the review argues, is that targeting one node invites compensation elsewhere. As the authors write, "This integrated perspective underscores that therapeutic interventions targeting a single pathway may be compensated for by others, highlighting the need for combination strategies."
A loop that runs in both directions
Much of the review is devoted to the crosstalk between mitochondrial oxidative stress and autoimmune responses, and it traces the loop in both directions. In one direction, oxidative damage converts mitochondrial DNA into oxidized mtDNA, which is highly immunogenic; it can be released into the cytoplasm through mitochondrial permeability transition pores and other channels, where it is recognised by the cGAS-STING pathway and induces a type I interferon response. Oxidized mtDNA can also be taken up and sensed by TLR9 in plasmacytoid dendritic cells, further amplifying interferon signalling and promoting T follicular helper cell differentiation. Separately, mitochondrial components including cardiolipin and oxidized mtDNA can themselves serve as autoantigens and drive the production of specific autoantibodies.
In the other direction, the immune response worsens the mitochondrial problem. Pro-inflammatory cytokines inhibit the electron transport chain and increase electron leakage: TNF-α activates protein kinase Cθ, which phosphorylates the NDUFS4 subunit of complex I and reduces its activity, while IFN-γ raises the expression of uncoupling protein 2 through the JAK-STAT pathway, lowering membrane potential and increasing proton leak that generates more ROS. Activation of NADPH oxidase in immune cells, oxidative stress that impairs the immunosuppressive function of regulatory T cells, and impaired antioxidant defence including reduced peroxiredoxin function all feed back into the same cycle. Oxidation of mitochondrial components in the central nervous system has been observed in the experimental autoimmune encephalomyelitis model before clinical symptoms appear, which the review presents as evidence that the process is not merely a downstream consequence of inflammation.
The review also maps how the loop reaches immune cell fate. Impaired mitophagy in T cells allows damaged, ROS-producing mitochondria to accumulate, pushing cells towards a pro-inflammatory Th17 phenotype while eroding the suppressive capacity of regulatory T cells. Metabolic reprogramming of autoreactive T cells towards glycolysis promotes the transcription factors that drive pro-inflammatory responses, and oxidative stress influences gene expression through DNA methylation and histone modification, including changes that favour effector T cell differentiation and function.
Three diseases, three mechanisms
The comparative analysis is where the review departs most clearly from a general account of oxidative stress. In rheumatoid arthritis, the hypoxic joint environment induces ROS accumulation and mitochondrial damage in immune cells and synovial fibroblasts. Synovial fibroblasts show a pronounced Warburg effect, with impaired oxidative phosphorylation and enhanced glycolysis, producing an imbalance in the NADH/NAD + ratio and increased electron leakage. Hypoxia-inducible factor 1α is persistently expressed and further amplifies glycolytic metabolism, and TNF-α signalling inhibits complex I directly. Oxidized mitochondrial DNA is elevated in synovial fluid, where it activates macrophages through TLR9. The resulting damage includes pannus formation driven by the ROS–HIF-1α–VEGF axis, cartilage degradation through NF-κB-induced matrix metalloproteinases, and chondrocyte apoptosis.
In systemic lupus erythematosus, the emphasis falls on defective mitophagy and interferon signalling. Post-translational modifications reduce the binding of mitophagy receptors to autophagy proteins in T cells, allowing damaged mitochondria to accumulate, and abnormal methylation patterns in the promoter regions of genes linked to mitochondrial biogenesis affect mitochondrial turnover. Oxidatively damaged mitochondrial DNA drives the type I interferon response through cGAS-STING, while plasmacytoid dendritic cells are unusually reactive to oxidized mtDNA. Oxidized mtDNA then disseminates to the kidneys, skin and central nervous system, and oxidative modification of autoantigens such as histone citrullination and DNA oxidation promotes autoantibody production.
In multiple sclerosis, the review points to energy failure in the cells that make myelin. Oligodendrocytes show reduced activity of electron transport chain complexes and substantially impaired ATP production, an imbalance in mitochondrial calcium homeostasis promotes abnormal opening of the permeability transition pore, and disrupted mitochondrial transport along axons causes uneven energy distribution. Falling NAD + levels impair deacetylase activity and weaken antioxidant defences. Integrity of the blood-brain barrier is compromised by mitochondrial ROS in endothelial cells and reduced expression of tight junction proteins, while microglial inflammasome activation and synaptic mitochondrial dysfunction contribute to axonal degeneration.
From mechanism to therapy, and where it stalls
The review surveys the therapeutic strategies that follow from this map: mitochondria-targeted antioxidants such as MitoQ and SS-31; natural compounds including resveratrol and its derivative pterostilbene; antioxidant nanoparticles, which offer advantages in stability and targeted delivery but remain largely unexplored in autoimmune disease; regulators of mitophagy; inhibitors of NOX4 and the NLRP3 inflammasome; interventions aimed at regulatory T cell function; and metabolic reprogramming through the AMPK/SIRT1/PGC-1α axis.
It also sets a boundary on how far that approach should go. ROS serve essential physiological signalling functions in immune cells, including T cell activation and differentiation, so a general strategy of indiscriminately eliminating them may impair normal immune surveillance and host defence. The review's stated preference is to restore redox balance rather than abolish ROS production, ideally by targeting pathological sources while preserving physiological signalling. The authors strike a similar note of caution about transplanting results between fields: evidence that regulating mitochondrial oxidative stress can suppress autoreactive CD4 + T cell subsets has come largely from cancer models, and direct translatability to autoimmune disease requires further investigation.
The obstacles to clinical translation, the review argues, are disease-specific. In rheumatoid arthritis, the challenge is reaching the inflamed synovium and penetrating relatively avascular cartilage. In systemic lupus erythematosus, it is the systemic nature of the disease, which increases the risk of off-target effects and requires a broad therapeutic window across multiple affected organs. In multiple sclerosis, agents must cross the blood-brain barrier to reach the central nervous system while minimising central nervous system side effects.
What still stands in the way
The review is candid about the gap between preclinical promise and clinical readiness. Clinical diagnostic tests for monitoring mitochondrial dysfunction and oxidative stress are lacking, and the molecular mechanisms by which mitochondrial oxidative stress affects cell fate remain unclear. In central nervous system autoimmune disease, the role of mitochondrial-regulated regulatory T cells is still undetermined. Peripheral blood biomarkers are a promising avenue for patient stratification, pharmacodynamic monitoring and trial design, and candidates include plasma oxidized mtDNA, which has been associated with disease activity in systemic lupus erythematosus, circulating cell-free mtDNA, mitochondrial ROS production in peripheral blood mononuclear cells, and serum markers of oxidative damage such as 8-hydroxy-2'-deoxyguanosine and malondialdehyde. The caveats attached to these markers are substantial: assays are not standardised, many oxidative stress markers are unstable outside the body, distinguishing mitochondrial from non-mitochondrial ROS is difficult, and validated cut-off values for clinical decisions are absent. As the authors write, "Furthermore, most studies have been cross-sectional, and longitudinal data linking biomarker changes to treatment response are sparse." They add that future trials should incorporate serial biomarker measurements to establish their utility in guiding therapy and predicting relapse.
Outlook
The review's forward-looking agenda follows from the network it describes. Single-cell multi-omics and organoid models are proposed to define disease-specific mitochondrial-immune networks and cell-type-specific vulnerabilities; nanocarriers are proposed to improve targeting; combination therapies with existing drugs are proposed to overcome compensatory signalling; and disease assessment systems are proposed that incorporate mitochondrial functional parameters rather than relying on biomarkers alone. As the authors write, "With continued research, mitochondrial-targeted therapies may offer new treatment options for autoimmune diseases."
Article information:
Mitochondrial oxidative stress in rheumatoid arthritis, systemic lupus erythematosus, and multiple sclerosis: From pathogenic mechanisms to therapeutic prospects.
Medicine Discovery , 2026.
Original article link: https://www.sciencedirect.com/science/article/pii/S3117928226000056
DOI: 10.1016/j.medd.2026.100005
About the journal
Medicine Discovery (ISSN: 3117-9282) is a peer-reviewed, international, open-access journal published by Zhujiang Hospital of Southern Medical University in collaboration with KeAi Publishing and distributed on the ScienceDirect platform. The journal is dedicated to publishing high-quality translational medical research that bridges mechanistic discovery and clinical implementation, and covers five sections spanning mechanisms and systems medicine, translational and clinical research, digital medicine and data-driven healthcare, therapeutic innovation and biomedical engineering, and methodology, standards, and implementation science.
Website: https://www.sciencedirect.com/journal/medicine-discovery
Submit: https://www.editorialmanager.com/medd/default2.aspx
Medicine Discovery
Literature review
People
Mitochondrial oxidative stress in rheumatoid arthritis, systemic lupus erythematosus, and multiple sclerosis: From pathogenic mechanisms to therapeutic prospects