Finding. Mitochondrial neurogastrointestinal encephalomyopathy is a thymidine-phosphorylase blank. Nucleosides pile up in blood, mitochondrial nucleotide pools skew, the genome frays, people die. Carrodus, van den Ameele and colleagues put human TYMP mRNA in lipid nanoparticles, inject it, and the mouse liver makes the enzyme. Circulating nucleosides fall to wild-type within hours and stay down as long as three weeks. The preclinical minimally effective dose is 0.25 mg/kg. The shots are tolerated. If you want a needle that is not a vein, subcutaneous LNPs work when you add hyaluronidase, either as protein or as another mRNA (SPAM1).
Why this paper matters
Today’s MNGIE options are liver or marrow transplants: real enzyme, real risk, real donor problem. An mRNA that turns the liver into a nucleoside sink for weeks is a different product. Transient expression is a feature. You do not need the liver to keep a transgene forever if you can redose.
It is also a platform sentence. Any monogenic disease whose missing protein can work from liver into plasma might use the same cassette, including the subcutaneous-plus-hyaluronidase trick.
How to read the score
Mid-nineties. Primary mitochondrial disease, a practical modality, a dose, a duration, a route. Confidence is high for the mouse nucleoside pharmacodynamics, not for human neuropathy.
Caveats
Nucleosides are not mtDNA copy number and not a clinical score. Immunity to repeat LNPs is the long-game risk. Do not skip transplant counseling from this brief.
What to do with it
If you track MDS therapeutics, this is the weekly mRNA paper. If you build liver-sink strategies, steal 0.25 mg/kg and the SPAM1 subcutaneous add-on. Pull the nucleoside time course before you write “cured MNGIE.”