An mRNA-rRNA Interaction Stimulates Frameshifting and Influences Initiation Accuracy
Early in our study of the +1 programmed frameshift site from the Ty3 element we identified a region immediately downstream of the frameshift site that stimulated frameshifting almost 10-fold. This 14 nt sequence, which we called the context, appears to function as a primary mRNA sequence. We explicitly showed that a putative pseudoknot that could be formed with sequences including the context was unimportant to stimulation. We also excluded the idea that the protein encoded by the context has any effect on frameshifting.
If the primary sequence stimulates frameshifting, how does it do it? The sequence of the context is complementary to a segment of the rRNA known to be involved in translational accuracy: the 530 loop of the small subunit rRNA. The 530 loop actually can adopt two conformations, a stem-loop and a pseudoknot. The structures are associated with occupancy of the A site: when a cognate tRNA occupies the A site the 530 loop is in the pseudoknot form, while when a non-cognate occupies it 530 is in the stem-loop form. Powers & Noller have suggested that the 530 loop is part of a mechanism that traps cognates in the A site, and allows dissociation of non-cognates.
Brimacombe's laboratory has also shown that the region the same distance downstream of the decoding site as the context during frameshifting crosslinks to the 530 loop. Complementarity of the context to the 530 loop should allow base pairing when the frameshift site enters the decoding center. We think that this base pairing may preclude the 530 loop from performing its error-correction role, which could cause an increase in +1 frameshifts. We are using combined genetic and biochemical approaches to study this issue.