Back

Cooperative Architecture of Mitochondrial Proteome Homeostasis

Forny, P.; Forny, M.; Smith, A. J.; Sung, A. Y.; Liu, K.; Pagliarini, D. J.

2026-02-09 genetic and genomic medicine
10.64898/2026.02.06.26345691 medRxiv
Show abstract

Mitochondria are semi-autonomous organelles whose generation and maintenance demand precise expression, processing, and assembly of >1,000 proteins encoded across two genomes. To explore this cooperativity, we performed multiomic analyses on >200 cell lines harboring mitochondrial gene perturbations, generating >26M molecular measurements. Our data reveal that mitochondrial proteome homeostasis is heavily influenced by post-transcriptional processes. Through nearest neighbor analyses, we reveal diverse protein activities undergirding this regulation, including MDH2s regulation of MT-ND3 transcription via FASTKD1 binding and CLPPs processing of the mitoribosomal assembly factor MALSU1, which we establish as a disease gene. Through entropy analysis, we reveal unexpectedly heterogeneous protein-level variability across complexes and use complexome profiling to identify new complex-specific membership, including C15orf61s association with complex V. We further observe substantial mtDNA copy number variation, notably upon disruption of the disease-related cobalamin biosynthesis protein MMADHC. Together, we establish new protein functions and provide a multilayered view into mitochondrial proteome regulation. HighlightsO_LIMultiomic signatures across perturbations reveal extensive post-transcriptional regulation C_LIO_LIThe TCA cycle enzyme MDH2 binds FASTKD1 to modulate MT-ND3 transcript levels C_LIO_LIMALSU1 is a CLPP protease substrate whose deficiency causes a mitochondrial disease C_LIO_LIC15orf61 binds ATP synthase and negatively regulates its higher order assembly C_LIO_LIMMADHC inversely affects mtDNA levels potentially mediated through LONP1 C_LI

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.