Finding. Pathogenic Mitofusin 2 (MFN2) variants cause Charcot-Marie-Tooth type 2A, a peripheral neuropathy. A subset of patients also get myopathy, and the fusion-GTPase textbook does not explain the muscle. Zaman, Shutt and colleagues report a conserved cellular phenotype for those myopathy variants: mitochondrial DNA leaves the organelle into RAB5 early endosomes and robustly activates Toll-like receptor 9 and cyclic GMP-AMP synthase / stimulator of interferon genes (cGAS/STING) signaling.
The leak is not one compartment. Mitofusin 2 dysfunction produces two extra-mitochondrial mitochondrial DNA pools. One sits in endosomes. The other rides mitochondrial-derived vesicles. You can tell them apart by size: endosomal mitochondrial DNA makes larger puncta. Turning on the integrated stress response with a drug shrinks both pools.
Why this paper matters
Fusion failure is a morphology problem. Endosomal mitochondrial DNA is an innate-immunity problem. Those are different therapeutic objects. If myopathy-linked Mitofusin 2 alleles dump genome into a RAB5 compartment that already owns Toll-like receptor 9, the muscle phenotype can be inflammation downstream of a trafficking error, not only a bioenergetic failure.
The two-pool split is the operational detail. Papers that stain "cytosolic mitochondrial DNA" will mix a vesicle shipment with an endosomal dump. Size and RAB5 identity let you unmix them. The integrated-stress-response result is the handle: one pharmacological axis reduces both pools in this system.
Patient fibroblasts are in the ethics statement, so this is not framed as a pure overexpression cartoon, even if the abstract withholds the allele table.
What they actually measured
The abstract gives a pathomechanism, two compartments, a size discriminant, two innate sensors, and a rescue class. It does not give the variant list, the integrated-stress-response compound, the puncta-size cut, or a muscle histology series. Treat "conserved" as a claim across the myopathy variants they tested, not across all Charcot-Marie-Tooth type 2A.
Keep neuropathy and myopathy distinct. Most Mitofusin 2 disease is length-dependent axon loss. This brief is about the myopathy-linked subset and a cellular inflammatory phenotype.
How to read the score
Nineties. Named disease gene, mitochondrial DNA outside the organelle, two compartments, two sensors, a drug class. Confidence is high for the cellular phenotype as stated. It is not a clinical trial and not a fusion-structure paper.
What to do with it
If you track mitochondrial DNA sensing, Mitofusin 2, or sterile inflammation in mitochondrial disease, split endosomal versus vesicle pools and score Toll-like receptor 9 alongside STING. Pull the patient-fibroblast panels and the integrated-stress-response rescue. Do not prescribe an integrated-stress-response agonist from this brief.