Chapter 19 : Recognition in the Glutamine tRNA System: from Structure to Function

The relaxed tRNA specificity of E. coli GlnRS applies to amber (anticodon CUA), not opal (anticodon UCA) suppressor tRNAs. Shown are the acceptor stems of the various tRNAs. While many amber suppressors (top two rows) or their mutants (bottom row) insert glutamine (%Q) in vivo, the opal suppressors derived from these same tRNAs, which have been tested, insert little (*) or no (**) glutamine in vivo. (Adapted from references 38 , 46 , 48 – 52 , 56 , 69 , 71 , 85 .)

Comparison of in vivo identity of amber (UAG) and opal (UGA) suppressors derived from tRNAGln. The amber suppressor supE, also known as glnV(SuUAG), is overdetermined for glutamine identity, because even multiple mutations in this context do not severely impair the glutamine specificity of these mutant tRNAs. On the other hand, the opal suppressor su +2op already inserts 88% tryptophan into dihydrofolate reductase. Additional mutations abolish GlnRS recognition. These mutant tRNAs are also probably not recognized well by other aaRS, as several of them are inactive suppressors. (Adapted from references 53 , 69 , and 85 .)

The effect of competition and tRNA context on the in vivo identity of amber suppressor tRNAs. In vitro, in its cognate tRNAGln context, GlnRS clearly prefers purine over pyrimidine discriminators ( 32 ). However, in vivo, glnV(SuUAG)-U73 inserts more glutamine into dihydrofolate reductase than the A/G73 variants, partially due to a lack of competition for glnV(SuUAG)-U73 ( 85 ). Similarly, of all the aaRS in E. coli, GlnRS, despite its tendency to favor purines in the discriminator position in tRNAGln, competes most effectively for the U73 mutants of the noncognate amber suppressors gluA and tyrT(SuUAG). Even more surprisingly, the A73 mutant of gluA is glutamate-specific, while gluA, with its wild-type discriminator is not ( 85 ). Clearly, the identity of gluA in vivo is determined by competition, with GlnRS competing less effectively for gluA-A73 than for gluA-G73.

Pathway for evolutionary divergence of the glutamate and glutamine aminoacylation systems. This model ( 68 ) is based on the preference of GluRS for U34-containing anticodons and on the fact that, in gram-positive eubacteria, archaebacteria, and organelles, GlnRS does not exist, and both glutamate and glutamine tRNAs have a U at position 34. GlxRS is an ancestral aaRS specific for both glutamate and glutamine. In gram-positive bacteria, duplication of the gene for GlxRS, coupled with divergence at position 36 of tRNAGlu and tRNAGln, may have led to the evolution of the Glu-tRNAGln amidotransferase and GluRS, with their respective glutamine and glutamate tRNA and amino acid specificities. In gram-negative bacteria, the creation by gene duplication and mutation of tRNAGlx CUG and duplication of the GlxRS gene is proposed to have led to co- evolution of GlnRS, with its C34-G36 specificity, and of tRNAGln. Dashed lines connect cognate tRNAs and aaRS and the broken arrow represents the possible divergence of the amidotransferase from a primitive GlxRS.