The Highly Conserved Codon following the Slippery Sequence Supports -1 Frameshift Efficiency at the HIV-1 Frameshift Site

Mathew, Suneeth F.; Crowe-McAuliffe, Caillan; Graves, Ryan; Cardno, Tony S.; McKinney, Cushla; Poole, Elizabeth S.; Tate, Warren P.

The first base of the intercodon could affect frameshift efficiency by stacking on the A site codon (UUA) at the end of the slippery sequence as the ribosome stalls at the stem-loop. This model predicts that a purine would stack more easily than a pyrimidine due to its larger size and, therefore, would lower frameshift efficiency by providing greater stability for frame maintenance. By contrast, a pyrimidine in the same position would tend to be less effective [54] . Our results with GGG versus UGA (Fig. 2) were the opposite to that expected if such a stacking mechanism were dominant. We tested this further with a series of frameshift constructs with all four bases (A, G, C, T) for the first position of the intercodon, while the third base was either in the NGG or the NGA format. The order of frameshift efficiency for the NGG series was UGG>GGG>CGG>AGG whereas for the NGA series it was GGA>CGA>AGA>UGA (Fig. 3 ). In particular, the two purine bases, G and A, resulted in substantially different levels of frameshifting, which would not be predicted if base stacking were responsible for the major effect of the intercodon. Intriguingly, the two first base U codons, UGG (12% frameshifting) and UGA (3.5% frameshifting) were at the extremes. Notably, UGG is decoded by a tRNA while UGA is decoded by the eukaryotic protein release factor eRF1.