“…We have previously suggested as one of a number of possibilities that the guide RNAs used in eukaryotes to identify the U residues destined for conversion to ⌿ might be RNA chaperones for the correct folding of ribosomal RNA (Ofengand & Fournier, 1998)+ In this view, ⌿ formation would be merely a signal that folding had occurred and it was time for the chaperone to dissociate, rather than having an intrinsic function+ If this view were extended to prokaryotes, substituting the synthases for guide RNAs, the ⌿ synthases might then be protein chaperones of RNA folding, helping in some as yet undefined way to achieve the correct structure+ However, it does not seem so likely that ⌿ formation is a necessary completion signal, as in its absence, when mutant RluD was used, complete growth rate restoration was observed (Table 2)+ More likely, RluD, and possibly the other synthases as well, have two distinct functions, one of which is related to the observed growth defects, and the other to ⌿ formation, the latter occurring for still unknown reasons+ There is some precedent for these ideas+ First, the methyltransferase that makes m 5 U54 in tRNA is indispensable in E. coli, yet its methylation activity is not required (Persson et al+, 1992)+ Second, Dim1p, the yeast enzyme which makes m 6 2 Am 6 2 A at the 39 end of the small subunit rRNA and is essential for the yeast cell (Lafontaine et al+, 1994), can dispense with its methylase activity (Lafontaine et al+, 1998)+ Third, two LSU rRNA 29-O-methyltransferases are known whose absence perturbs ribosome assembly and results in severe or lethal growth defects+ One is PET56, a yeast 29-O-methyltransferase specific for G2251 (E. coli numbering) in yeast mitochondria (Sirum-Connolly & Mason, 1993)+ The other is FtsJ, which makes Um2552 in E. coli (Bügl et al+, 2000;Caldas et al+, 2000aCaldas et al+, , 2000b)+ In the latter example, it is not known whether the 29-O-methylation is required for proper ribosome assembly, but in the former case, recent work has shown that rescue of ribosome assembly does not require methylation (T+ Mason, pers+ comm+)+ Thus, like RluD and TruB, the protein is needed but not the product of the reaction it catalyzes+ Inhibition of 50S subunit assembly in both Dust and Tiny RluD-minus strains has also recently been observed (L+ Peil, N+ Gutgsell, J+ Ofengand, & J+ Remme, unpubl+ results)+ Considering all these results, a pattern begins to emerge in which rRNA modifying enzymes, at least the most common ones, ⌿ synthases and 29-O-methyltransferases, may have an assembly function in their own right, independent of their catalytic role in modified base formation+ In this regard, it is interesting to note the amino acid sequence homology between the N-terminus of RluD (residues 19-49) of this 326 amino acid protein with residues 10-40 of Hsp15, a protein highly induced by heat shock (Korber et al+, 1999)+ Hsp15 is a 133 amino acid protein that consists almost entirely of a new type of RNA-binding fold and which binds to 50S subunits carrying nascent protein chains with nanomolar affinity + In this context, the occurrence of pseudorevertants should be noted+ Such second-site mutants are found readily in the Dust strain, and pains must be taken to avoid them during subculturing+ They occur much more rarely in the Tiny strain, possibly because the grow...…”