Multi-color single molecule imaging uncovers extensive heterogeneity in mRNA decoding

Sanne Boersma, Deepak Khuperkar, Bram M.P. Verhagen, Stijn Sonneveld, Jonathan B. Grimm, Luke D. Lavis, Marvin E. Tanenbaum

Preprint posted on November 24, 2018

When ribosomes take an alternative path: single-molecule sensor of translation reveals extensive heterogeneity in mRNA decoding

Selected by Lorenzo Lafranchi


Translation, the process by which a ribosome reads an mRNA molecule, is a crucial and tightly regulated step in gene expression. The canonical view of translation is that a ribosome scans the mRNA from its 5’ end, beginning protein synthesis at the most upstream start codon, and continues translation until encountering an in-frame stop codon. The region enclosed between in-frame start and stop codons is referred to as open reading frame (ORF). This simplistic view of translation has been recently challenged by the development of ribosome profiling, which highlighted the heterogeneity existing in mRNA decoding. In fact, mRNA molecules can contain multiple ORFs and different sections of an mRNA can be translated. Altogether, non-canonical translation has been widely observed but the extent and underlying causes of heterogeneity in mRNA translation remain largely unexplored.


Key findings

Boersma and colleagues develop a fluorescence reporter, called MoonTag, for labeling nascent polypeptides. Shortly, MoonTag consists of a genetically-encoded antibody-epitope pair. A short peptide is fused as an array to a sequence of interest and upon translation is recognized and bound by the fluorescently-labeled nanobody. Live-cell imaging then enables the analysis of translation kinetics. With the MoonTag being orthogonal to the previously-published SunTag system, the authors are now able to combine the two reporters for studying translational heterogeneity at a single-molecule level.

After testing the MoonTag and proving its orthogonality to the SunTag, the authors combine the two reporters in an alternating fashion and in different reading frames into the MashTag. Since the MashTag is provided with a single start codon, in-frame with either the Sun- or the MoonTag, canonical translation generates only one of the two fluorescent signals. The observed colocalization of SunTag and MoonTag signals therefore represents ribosomes that are translating the reporter out-of-frame (OOF).To further understand OOF translation, the authors compute the theoretical trace expected from a single ribosome translating the entire array of peptides. Comparison of the experimental profiles of single OOF translation events to the theoretical trace revealed that OOF translation is mainly due to alternative start site selection near the 5’ end of the mRNA. Next, the theoretical trace was used to extract information from the measured fluorescence intensity traces about the frequency and timing of translation initiation at both canonical and alternative start sites. This analysis showed that both canonical and OOF translation occur intermittently on the majority of mRNAs. On the individual mRNA level, start site selection seems to be largely stochastic. Surprisingly, the probability of using alternative translation start sites differs between mRNAs, with OOF translation ranging from 0% to 100% of the ribosomes. These findings highlight how different mRNAs, despite carrying the same ORFs, can be heterogeneously read by the ribosomes. Further experiments show that near-cognate translation start sites both upstream or downstream the canonical start codon can be responsible for initiating OOF translation.By fusing the 5’UTR of two genes to the MashTag, the authors demonstrate that OOF translation is not restricted to exogenous sequences and suggest that alternative start site selection might be, to different extents, a widespread phenomenon on endogenous mRNAs.

Finally, the authors design a sensor for visualizing the translational paths occurring on an mRNA containing an upstream ORF (uORF). uORFs are present in thousands of mRNAs and generally repress translation of the main ORF. Different to the previously-described reporters, the uORF sensor contains two out-of-frame canonical start codons. As expected, presence of the uORF reduces the translation rate of the main ORF, but translation of both ORFs occurs for most mRNA molecules. Translation of the main ORF can both be achieved by leaky scanning of the first start codon or translation re-initiation after translation of the uORF. Interestingly, ribosomes following different paths along the mRNA co-exist on most mRNAs.In addition, the authors observed a strong temporal correlation between translation of the two ORFs that is most likely caused by a burst-like behavior of translation initiation. Despite the majority of mRNAs showing a positive correlation between translational level of the two sensors, a fraction of mRNA molecules experiences temporal bursts in choosing the different translation paths. Bursts in translation start site selectionoccur in a mRNA-specific fashion indicating that this phenomenonis not regulated in a cell-wide manner, but at the level of individual mRNA molecules.


What I like about this work

Ribosome profiling and proteogenomics studies suggested that larger portions of the genome are translated in comparison to what was expected based on the textbook view of translation. These methods highlighted an unforeseen proteome diversity, possibly due to the flexibility of mRNA-translating ribosomes. Despite these findings, translational heterogeneity has never been directly visualized. Boersma and colleagues developed an elegant strategy to dissect the complexity of translational dynamics at a single-molecule level. With these new tools in hand they answer different open questions on how ribosomes initiate translation and they show that alternative translation is a surprisingly common event.


Future directions

Although the system presented in the paper seems difficult to multiplex, it would be interesting to fuse the MashTag to several endogenous 5’ UTRs; similar to what the authors already did in the paper, but with larger numbers. Collecting more information would help in defining sequences prone to OOF translation.

“What is the role and importance of OOF translation for cellular physiology?” is an important question arising from the study. OOF translation could be important for controlling translation of the main ORF, generate functional alternative proteins, or simply represent errors in translational initiation. Since translation is energetically costly for a cell, the first two explanations seem to be the most likely. Nevertheless, this open question is worth to be addressed experimentally.


Questions to the authors

How stable are the products of the MashTag? Is there a difference between “canonical” product and OOF-translated products?

Would it be possible to perform ribosome profiling after harringtonine treatment to identify the alternative initiation sites?

Can the overall level of OOF translation in a cell be measured using the MashTag reporter? Can this system be implemented in a genetic or chemical screen to identify factors controlling OOF translation?


Posted on: 18th December 2018

Read preprint (No Ratings Yet)

  • Have your say

    Your email address will not be published. Required fields are marked *

    This site uses Akismet to reduce spam. Learn how your comment data is processed.

    Sign up to customise the site to your preferences and to receive alerts

    Register here