Mitochondrial ribosome mutations recur in thousands of tumors and impair respiration at low doses

Mitochondria carry their own small genome, and cancers frequently mutate it. Until now, the mitochondrial ribosomal RNAs (12S, MT-RNR1; 16S, MT-RNR2) were not known to be selected for mutation in tumors. Germline mt-rRNA mutations are also rarely linked to disease: only 2 of 94 pathogenic variants in the MITOMAP database, both tied to aminoglycoside-induced hearing loss, sit in rRNA genes, versus 50 of 94 in tRNAs. Mitochondrial DNA mutations have also been considered functionally recessive, with clinical effects usually appearing only above roughly 60% heteroplasmy (the fraction of mtDNA copies carrying the mutation).

The researchers analyzed 14,106 primary tumors with whole-genome sequencing from the Genomics England 100,000 Genomes Project. Mean mtDNA coverage was about 15,919×, allowing confident calls down to about 5% heteroplasmy. They identified 18,104 somatic SNVs and 2,222 indels, and applied a hotspot-detection algorithm that tests for mutation burden above background mutational processes. This yielded 138 significant SNV hotspots across 21 tumor lineages, plus seven indel hotspots at homopolymeric sites in complex I genes. Of the SNV hotspots, 96 were in protein-coding genes, 8 in tRNA genes and 34 in rRNA genes.

The rRNA genes ranked second only to MT-ND5 for the number of hotspots, and about 4% of all tumors carried at least one rRNA hotspot. Only 7 of 138 SNV hotspots were also found in non-neoplastic cells, suggesting most are cancer-specific. Recurrence of mutant alleles in two independent cohorts, TCGA (exome) and PCAWG (whole genome), correlated with the Genomics England data (R = 0.703 and R = 0.94), which argues against a cohort or variant-calling artifact. According to the abstract, rRNA hotspots preferentially fell on positions under purifying selection in the germline and clustered structurally within the mitoribosome at mRNA- and tRNA-interacting positions.

To test function, the authors used precision mtDNA base editing to model an exemplar MT-RNR1 hotspot, m.1227G>A. Multimodal profiling showed a heteroplasmy-dependent decline in mitochondrial function and loss of respiratory chain subunits from a dosage of about 10%. This is far below the thresholds seen in classical mitochondrial disease and indicates that disruption of conserved rRNA positions can act in a functionally dominant way. The authors propose these mutations as a class of pathogenic mtDNA variants under positive selection in cancer.

These are genomic and cell-based findings, not clinical guidance. The text available for this summary was truncated partway through the Results section, so it does not cover how the mutations affect tumor growth, the full Discussion, or the details of the base-editing models beyond the abstract. Whether rRNA mutations drive tumor progression or create therapeutic vulnerabilities remains to be shown.