Ribosomal protein S12 undergoes a unique posttranslational modification, methylthiolation of residue D88, in Escherichia coli and several other bacteria. Using mass spectrometry, we have identified the enzyme responsible for this modification in E. coli, the yliG gene product. This enzyme, which we propose be called RimO, is a radical-S-adenosylmethionine protein that bears strong sequence similarity to MiaB, which methylthiolates tRNA. We show that RimO and MiaB represent two of four subgroups of a larger, ancient family of likely methylthiotransferases, the other two of which are typified by Bacillus subtilis YqeV and Methanococcus jannaschii Mj0867, and we predict that RimO is unique among these subgroups in its modification of protein as opposed to tRNA. Despite this, RimO has not significantly diverged from the other three subgroups at the sequence level even within the C-terminal TRAM domain, which in the methyltransferase RumA is known to bind the RNA substrate and which we presume to be responsible for substrate binding and recognition in all four subgroups of methylthiotransferases. To our knowledge, RimO and MiaB represent the most extreme known case of resemblance between enzymes modifying protein and nucleic acid. The initial results presented here constitute a bioinformatics-driven prediction with preliminary experimental validation that should serve as the starting point for several interesting lines of further inquiry.methylthiotransferase ͉ posttranslational modification ͉ radical-SAM protein T he translational apparatus undergoes numerous posttranscriptional and posttranslational modifications of its RNA and protein components, respectively, and the enzymes responsible for many of these modifications have been identified in recent years (1, 2). One modification for which the responsible enzyme is currently unknown is the methylthiolation of the -carbon of residue D88 of ribosomal protein S12 in Escherichia coli (Fig. 1A) (3). Intriguingly, D88 is strictly conserved in all S12 homologs from bacteria, archaea, and eukaryotes. No spontaneous D88 mutants have been identified, and attempts to alter the residue by site-directed mutagenesis have failed where mutation of surrounding residues succeeded (4), suggesting that this aspartic acid residue serves an essential function. The methylthiol modification has also been observed in S12 from Rhodopseudomonas palustris and Thermus thermophilus (3,5,6), but it appears not to be present in cytoplasmic ribosomes from rat and human cells (7,8). In T. thermophilus, S12 P90R and P90W mutants have been shown to be deficient in the D88 modification, presumably due to steric hindrance of the modifying enzyme by the bulky sidechains (4). Thus, although D88 itself appears to be universal and essential, the methylthiol modification is not.The function of this modification remains unclear, but its location suggests a possible role in decoding or translocation. The solved crystal structure of S12 from T. thermophilus reveals that D88 sits within one of two highly cons...