https://www.cell.com/cell-reports-medicine/fulltext/S2666-3791(26)00258-2
chatGPT:
Summary
This is a 2026 review article on emerging therapies for primary mitochondrial diseases caused by mitochondrial DNA defects. Its central argument is that mitochondrial medicine is moving from supportive care toward direct mtDNA-targeted intervention.
The paper starts from the biological problem: mtDNA diseases are difficult because mtDNA is multicopy, often heteroplasmic, variably distributed across tissues, maternally inherited, and physically hard to access inside the mitochondrial matrix. The authors emphasise that conventional CRISPR-Cas approaches are not straightforward because mitochondria lack an efficient endogenous route for importing guide RNA.
The review then groups therapeutic strategies into three broad classes:
-
Gene therapy approaches
- Allotopic expression: nuclear expression of a recoded mitochondrial gene, followed by import of the protein into mitochondria. The leading example is lenadogene nolparvovec / GS010 / LUMEVOQ for LHON caused by the MT-ND4 m.11778G>A mutation. This is the most clinically advanced mtDNA-targeted gene therapy discussed.
- Programmable nucleases: mitoREs, mtZFNs, mitoTALENs and mitoARCUS. These are designed to selectively cut mutant mtDNA, leading to degradation of mutant genomes and a shift in heteroplasmy toward wild-type mtDNA.
- Base editors: DdCBEs, TALEDs, mitoBEs and related systems, which aim to correct pathogenic point mutations without double-strand breaks. This is presented as especially important because many mtDNA diseases are caused by point mutations.
-
Pharmacological heteroplasmy modulation
- The paper reviews small molecules that may shift heteroplasmy by affecting mtDNA replication, mitochondrial quality control, metabolism, or mutant mtDNA disadvantage.
- Examples include berberine/RHPS4 for G-quadruplex-associated mutant mtDNA, CDDO/bardoxolone-related mechanisms, PI3K-AKT-mTORC1 inhibitors including rapamycin, and glucose analogues such as 2-deoxyglucose and 5-thioglucose.
-
Reproductive and mitochondrial transfer strategies
- Preimplantation genetic testing can select embryos with low mutant heteroplasmy.
- Mitochondrial donation / replacement therapy includes maternal spindle transfer, pronuclear transfer and polar body transfer.
- Mitochondrial transplantation or augmentation is discussed as a therapeutic rather than reproductive approach, especially in cell-based settings such as hematopoietic stem-cell augmentation for single large-scale mtDNA deletion syndromes.
The paper concludes that the field is at a turning point: allotopic expression has reached phase III clinical testing, newer editors make mtDNA manipulation increasingly realistic, mitochondrial donation has produced live births, and pharmacological approaches are beginning to move beyond general mitochondrial support.
Novelty
The paper’s novelty is not that it reports a new experiment; it is a synthesis review. Its useful contribution is that it brings together several fast-moving areas that are often discussed separately:
- mtDNA nucleases and base editors;
- allotopic expression;
- small-molecule heteroplasmy modulation;
- mitochondrial donation;
- mitochondrial transplantation.
A particularly useful feature is its attempt to map specific diseases, mutations and candidate therapeutic approaches in Table 1. For example, it links MELAS m.3243A>G to mitoARCUS, mitoTALENs, PI3K/AKT/mTOR inhibitors and glucose analogues; LHON m.11778G>A to allotopic expression, mitoABE, TALEDs and rapamycin; and single large-scale deletion syndromes to mtZFNs, mitoTALENs, CDDO and mitochondrial augmentation.
The most genuinely current part is the coverage of next-generation mitochondrial base editing. The authors frame 2025 advances in delivery and editing systems as moving mtDNA gene therapy from proof-of-principle toward translational plausibility.
Critique
The review is strong as a broad landscape paper, but it is somewhat optimistic. Many of the approaches are still mainly cell or animal proof-of-principle, and the paper sometimes treats “can edit mtDNA” as closer to “can become a therapy” than is justified. The hard problems remain delivery to affected human tissues, dose control, immune response, long-term safety, off-target editing, tissue heterogeneity and durable correction.
The paper also mixes very different evidential levels. For example, LHON allotopic expression has reached clinical trials, whereas many nuclease and base-editor approaches remain preclinical. Small molecules that shift heteroplasmy in cybrids or fibroblasts are not in the same evidential category as reproductive mitochondrial donation with live births or an AAV-based clinical programme. The tables are helpful, but they risk making the field look more therapeutically mature than it is.
A second weakness is that heteroplasmy is treated mainly as a therapeutic target, but the biological complexity is deeper. Heteroplasmy varies by tissue, changes with age, and may not be well represented by blood, urine or fibroblast measures. The review does acknowledge tissue variability and threshold effects, but it could have gone further in explaining how this complicates trial endpoints, patient selection and monitoring.
A third limitation is the lack of a strong comparative framework. The paper describes many technologies, but does not clearly rank them by:
- likely clinical tractability;
- best disease fit;
- delivery feasibility;
- reversibility;
- off-target risk;
- scalability;
- regulatory difficulty.
For example, an intravitreal AAV therapy for LHON is a much easier delivery problem than correcting heteroplasmic mtDNA mutations in brain, skeletal muscle or heart. That distinction could have been made more sharply.
The discussion of pharmacological approaches is interesting but needs caution. Rapamycin, PI3K inhibitors, glucose analogues and replication-modifying agents may shift heteroplasmy under some experimental conditions, but systemic manipulation of metabolism or mtDNA replication could have complex effects in patients with already compromised mitochondrial function. The review notes safety issues in places, but a more critical treatment of dose, tissue specificity and long-term risk would strengthen it.
Overall, this is a useful, up-to-date review of a rapidly changing field. Its main value is as a map of emerging therapeutic strategies for mtDNA disease. Its main weakness is that it compresses a wide range of speculative, preclinical and clinical interventions into a single optimistic translational narrative. The field is clearly moving forward, but most mtDNA-targeted therapies still face a large gap between molecular feasibility and safe, durable clinical benefit.