tRNA-like Structures in Viral RNA Genomes

Abstract

Most of the eukaryotic mRNAs and a great number of genomes from plus-stranded RNA viruses contain at their 3′ ends a poly(A) sequence of variable length. All known prokaryotic mRNAs and bacteriophage RNAs as well as a few eukaryotic mRNAs are devoid of such a sequence. Among plant RNA viruses, the genomes not possessing this poly(A) stretch are fairly numerous. Of the 16 or so groups of plant RNA viruses (Wildy 1971), in only one group does the RNA terminate by poly(A).

In all known cases of mRNAs or of viral RNAs, whether or not they contain poly(A), the coding regions are followed by an untranslated heteropolymeric sequence. In human and rabbit β-globin mRNAs, the poly(A) stretch is preceded by a sequence of 134 and 95 untranslated nucleotides, respectively (Proudfoot 1977). The sea urchin mRNA for histone H4, which does not possess poly(A), presumably has an untranslated region at its 3′ end of about 30 nucleotides (Grunstein and Grunstein 1978). In the RNA of the bacteriophage MS2, the length of the 3′ untranslated sequence that follows the replicase gene is 174 nucleotides long (Fiers et al. 1976); this RNA, as well as the genomes of several other RNA bacteriophages, terminates with the sequence CCA identical to that of tRNAs.

The physiological function of these nontranslated regions at the 3′ end of mRNAs has not been completely elucidated. It has been reported that the poly(A) stretch protects the mRNAs, probably against 3′ → 5′ exonucleases and thus confers greater stability to...