In the workshop where ribosomes are assembled, the molecular tailor YbeY stitches an extra link into the protein chain of uS11. That link is isoaspartate, an amino acid whose backbone is one methylene group longer. The insertion makes the protein slightly longer, and it settles more tightly into the ribosome's framework. This conversion was once considered a random defect that accumulates with age in the proteins of the eye's lens. But in uS11, isoaspartate is present in all organisms, from bacteria to humans, and always at the same spot.
The problem was speed. E. coli divides every 20 minutes, while the spontaneous rearrangement of asparagine to isoaspartate takes about a day and a half. So the cell needs an accelerator. The authors scanned all 4403 proteins of the E. coli proteome with the AlphaFold3 neural network to find one capable of binding uS11 in the right pose. Of all the proteins, only three produced a confident model, and just one, YbeY, placed asparagine-119 next to a zinc ion.
Next, biochemists tested YbeY in a test tube. Within an hour, the enzyme converted almost all of uS11 into the isoaspartate form. Mutations that prevent YbeY from holding zinc or the substrate reduced its activity. But a mutation that impairs YbeY's other job, cutting RNA, had no effect on the link insertion. So the protein has two separate tools: one for RNA, another for proteins.
Archaea and eukaryotes lack YbeY. There, the same job is probably done by the ATPase Fap7. In the crystal structure of the Fap7-uS11 complex from the archaeon Pyrococcus abyssi, the authors saw that residue 124 is actually isoaspartate, not aspartate as previously thought. It seems evolution independently invented the same way of fine-tuning a ribosomal protein twice.
Why does the cell need this modification? When the ybeY gene was deleted in E. coli, ribosomes became worse at holding onto the protein bS21. The density of bS21 on the cryo-electron microscopy map decreased, and the contact between arginine-35 of bS21 and isoaspartate-119 of uS11 disappeared. Such ribosomes assembled less efficiently into complete 70S particles and translated a reporter protein at only 30% of normal.
Now it is clear why this modification matters so much. The tailor YbeY stitches in the extra link within minutes, and without that insertion the ribosome falls apart. Understanding how YbeY works could aid antibiotic development: the enzyme is conserved in bacteria but absent from human cytoplasm. We can target it without harming our own ribosomes.
The spontaneous conversion of asparagine to isoaspartate in a peptide with the Asn-Gly motif has a half-life of about 1.2 days, but YbeY speeds this reaction up to minutes, which is necessary for assembling the ribosome within the E. coli cell cycle (about 20 minutes).
Not published on arXiv. Here are its page, the PDF and the formalization: you can check it yourself, not only read our account.