Live-cell single RNA imaging reveals bursts of translational frameshifting

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

Although frameshifting has been extensively studied in vitro and in bulk assays (Caliskan et al., 2014; Charbonneau et al., 2012; Chen et al., 2014; Lopinski et al., 2000; Mouzakis et al., 2013; Ritchie et al., 2017) , the process has never been observed at the single molecule level in living cells. This leaves many basic questions about frameshifting unresolved. In particular, it is not clear how heterogenous frameshifting is from one RNA to another, nor is it clear if single RNA continuously frameshift in a constitutive fashion or if instead they frameshift in definable bursts, as has been observed for transcription (Lionnet and Singer, 2012) and 0-frame translation (Wu et al., 2016) . Finally, the localization of frameshifting has never been investigated, so it is not clear if frameshifting occurs all throughout the cell or is instead preferentially localized to specific sub-cellular regions. In the case of HIV-1 gag-pol, for example, previous assays have shown that 5-10% of translated protein product is frameshifted (Brierley and Dos Ramos, 2006; Dulude et al., 2002; Grentzmann et al., 1998; Mouzakis et al., 2013) . This is thought to occur when ribosomes translate through a specialized frameshift sequence (FSS) containing a stem loop structure that slows down incoming ribosomes and causes them to slip back one nucleotide on a slippery sequence preceding the stem loop. Whether or not this occurs constitutively and with equal probability on all HIV-1 RNA or if instead it occurs on a specialized subset that are in the right place, at the right time, and with the right factors remains to be determined.