Live-cell single RNA imaging reveals bursts of translational frameshifting

Kenneth Lyon; Luis U. Aguilera; Tatsuya Morisaki; Brian Munsky; Timothy J. Stasevich

The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. . https://doi.org/10.1101/478040 doi: bioRxiv preprint Because the estimated average pause time, 1/ * , is greater than the average initiation time, 1/ , ribosomes could initiate faster than they clear the FSS with an excess rate of − * = 0.009 per second and create upstream traffic jams in frameshifting sites. These traffic jams would continue to build for as long as the RNA remains in the frameshifted state, or approximately 1/ = 770 s on average. This would allow for the accumulation of approximately ( − * ) / ~ 7 ribosomes (on the HA multi-frame tag, Table 1 ). Occasionally, traffic jams can extend all the way back to the start codon, as seen in a sample simulation (Movie S11). However, because these traffic jams occur upstream of the FSS within the HA epitope repeats, the nascent chain intensity at translation sites is only partially correlated with ribosomal occupancy. As a result, the model predicts frameshifting sites are slightly brighter than non-frameshifting sites ( Fig. 5G ), which we also observe experimentally (Fig. S15 ). Moreover, due to the long time it can take to clear traffic jams, frameshifting can persist for hours following the global shut down of translation initiation (with harringtonine, for example), as can be seen in simulations of the best-fit model (Movie S12) and consistent with what we observed in Figs. 4C and S11. Thus, the final bursty model suggests a mechanism by which frameshifting can persist for long periods in the absence of translation initiation.