Changing views of translation: from ribosome profiling to high resolution imaging of single molecules in vivo

There are several mRNA surveillance pathways in eukaryotes (NGD, NSD and NMD) that moderate the effects of natural errors in the cell and more broadly regulate gene expression. Dr Green has previously defined biochemical parameters of the factors Dom34, Hbs1 and Rli1 in her in vitro reconstituted yeast translation system. She has correlated these biochemical observations with ribosome profiling experiments in yeast to broadly define the in vivo targets of these same mRNA surveillance pathways. In the process of studying these pathways, she has learned that different ribosome footprint (RPFs) sizes represent distinct states of the ribosome in a cell, and that these signatures can inform us on the molecular stresses that the ribosomes encounter. She is currently defining how various stresses encountered by cells trigger specific translational distress, and how the cell responds to these molecular triggers. Other studies in mammalian platelets and reticulocytes revealed the accumulation of ribosomes in the 3’ UTR of mRNAs and we connected this phenomenon with diminished levels of ribosome rescue (DOM34/HBS1L) and recycling (ABCE1) factors in these systems. She is currently drawing together related observations in the yeast and mammalian systems to better define the factors and mechanisms that maintain cellular ribosome homeostasis and to explore how it impacts development and stress-responses.

Dysregulation of protein synthesis is a common theme in many neurological disorders. In tuberous sclerosis complex, alterations in TSC cause mTOR, a master regulator of translation, to become hyperactivated. Fragile X syndrome involves loss of function of the polysome-associated RNA-binding protein FMRP, which is thought to regulate translation elongation. Professor Sim's developed a new, high-sensitivity ribosome profiling technique and used it for genome-wide analysis of protein synthesis in animal models of these two neurological disorders. His analysis reveals new insights into the molecular mechanisms through which the associated genetically altered proteins impact protein synthesis in the brain.

Regulation of protein synthesis on mRNA is central to eukaryotic expression of genes. Regulatory inputs are specified by the mRNA untranslated regions (UTRs) and often target translation initiation. Initiation consists of dynamic and complex interactions between eukaryotic initiation factors (eIFs) and the small ribosomal subunit (SSU) and concludes with joining of the large ribosomal subunit (LSU) to form a full ribosome and translate the mRNA code. While control of protein synthesis is important for fast-paced cell adaptation, methods to study the mechanistic intricacies of translation initiation in vivo transcriptome-wide were lacking.

A translation complex profile sequencing (TCP-seq) has been developed; a method related to the ribosome profiling approach. TCP-seq uniquely allows to resolve all translation intermediates, including the elusive mRNA ‘scanning’ by SSUs, visualize start codon recognition events, and capture diverse conformations of elongating ribosomes in vivo. Thus, the method can be used to pinpoint mRNAs with high potential for functional control and visualize locations of the regulatory elements in their 5'UTRs.

Combining TCP-seq with eIF-selective purification of complexes, Professor Preiss now aims to discern the fundamental mechanistic problems of initiation, such as what proteins confer directionality to the SSUs during scanning, and how ribosomal recruitment to mRNA is controlled during nutrient stress conditions. He will further adapt TCP-seq to use in mammalian cells and aim to obtain snapshots of translation in normal and cancer cells.

Professor Bakau reported the dissection of the flux of newly synthesized proteins through the system of co-translationally engaged factors in E. coli and S. cerevisiae using ribosome profiling technology. In E. coli, the SRP targets inner membrane proteins to the membrane, whereas the chaperone trigger factor (TF) associates primarily with cytosolic, periplasmic and outer membrane proteins, indicating nascent chains are triaged between SRP and TF pathways. The Hsp70 chaperone DnaK interacts in concerted action with TF, typically with longer nascent chains and reflecting domain boundaries of the emerged nascent chains. The assembly of protein complexes initiates co-translationally once the domain interfaces of the interacting polypeptides are exposed. The organisation of the subunit-encoding genes in operons facilitates the assembly process. In S. cerevisiae the Hsp70 chaperone Ssb shows a different pattern of early nascent chain association as compared to its prokaryotic DnaK counterpart. Co-translational protein assembly is also observed, despite operon organisation of genes is less common, pointing to differences in the principles governing protein assembly in pro- and eukaryotes.

Professor Stasevich has developed technology to image single RNA translation dynamics in living cells. Using high-affinity antibody-based probes, multimerized epitope tags, and single molecule microscopy, we are able to visualize and quantify the emergence of nascent protein chains from single pre-marked RNA1. In this talk, he will describe this technology as well as a new multi-frame tag that extends the technology to enable real-time quantification of single RNA translation kinetics in any two of the three possible open reading frames. As a first application of the multi-frame tag, he uses it to dissect the kinetics of the HIV-1 frameshift sequence. Whereas previous bulk assays have shown this sequence leads to ~10% frameshifted product, it was not clear if all RNA frameshift with ~10% efficiency or if instead ~10% of RNA frameshift with ~100% efficiency. Interestingly, our live-cell data suggest the latter scenario, where a small subset of genetically identical RNA frameshift with high efficiency. The origin of this heterogeneity is not yet clear, but experiments are beginning to implicate a ribosomal pause that leads to a higher than normal density of ribosomes near the frameshift sequence on frameshifting RNA.

After transcription, an mRNA's fate is determined by an orchestrated series of events (processing, export, localisation, translation and degradation) that is regulated both temporally and spatially within the cell. In order to more completely understand these processes and how they are coupled, it is necessary to be able to observe these events as they occur on single molecules of mRNA in real-time in living cells. To expand the scope of questions that can be addressed by RNA imaging, we are developing multi-color RNA biosensors that allow that status of a single mRNA molecules (e.g. translation or degradation) to be directly visualised and quantified.  

In order to image the first round of translation, Dr Chao has developed TRICK (translating RNA imaging by coat protein knock-off) which relies on the detection of two fluorescent signals that are placed within the coding sequence and the 3′UTR.  In this approach, an untranslated mRNA is dual labeled and the fluorescent label in the coding sequence is displaced by the ribosome during the first round of translation resulting in translated mRNAs being singly labeled. A conceptually similar approach was used for single-molecule imaging of mRNA decay, where dual-colored mRNAs identify intact transcripts, while a single-colored stabilized decay intermediate marked degraded transcripts (TREAT, 3′ RNA end accumulation during turnover). Dr Chao is using these tools to characterise localised translation and degradation during normal cell growth and stress. 

Guided by a small RNA, Argonaute 2 (Ago2) interacts with its target mRNA, resulting in mRNA cleavage and decay. In vitro measurements have provided important insights in the binding and cleavage kinetics of Ago2, but much less is known about the behavior of Ago2 in living cells. Here, Marvin describes a method based on his previously developed SunTag translation imaging system, which allows him to observe cleavage of single mRNAs by Ago2 in living cells. He found that Ago2 frequently cleaves its target mRNA within minutes after nuclear export, at a time that precisely coincides with the arrival of the first translating ribosome at the Ago2 binding site. Using translation drugs and different mRNA reporters, he showed that ribosomes can stimulate mRNA decay by Ago2, by promoting the release of cleaved mRNA fragments from Ago2. However, the role of ribosomes in modulating Ago2 activity is paradoxical, as ribosomes also inhibit mRNA cleavage by Ago2, by displacing Ago2 molecules from the mRNA before cleavage can occur. Whether ribosomes promote or inhibit cleavage depends on mRNA release kinetics of the small RNA/Ago2 complex, and Ago2 showed distinct release kinetics when associated with different small RNAs. In summary, through live-cell single molecule imaging, he found that ribosomes profoundly alter Ago2’s mRNA cleavage activity by modulating distinct rate constants of the Ago2 cleavage cycle.

At least half of human genes use alternative cleavage and polyadenylation to generate mRNA transcripts that differ in the length of their 3' untranslated regions (3'UTRs) while producing the same protein.  We found that 3'UTRs can mediate protein-protein interactions, and thus, can determine protein localization and protein functions. Dr Mayr showed that the long 3'UTR of CD47 is required for the formation of a protein complex between CD47 and SET. This interaction enables CD47 protein to efficiently localise to the cell surface, but it also changes protein function as only CD47 protein that was generated from the long 3'UTR isoform (CD47-LU) is able to activate the Ras GTPase RAC1.  

Dr Mayr investigated the mechanism of 3'UTR-dependent interaction between SET and CD47 and found that the RNA-binding protein TIS11B is necessary for this process. TIS11B is known to bind to AU-rich element (ARE) containing mRNAs and destabilizes specific mRNAs. Using super-resolution live cell imaging, we observed that TIS11B assembles into a large tubule-like meshwork that is intertwined with the endoplasmic reticulum (ER). The TIS11B granules enrich membrane protein-encoding mRNAs that contain multiple AREs in their 3'UTRs, including CD47-LU, and they exclude mRNAs lacking these features. The TIS11B granules also enrich specific proteins including chaperones, but they exclude SET.  The spatial arrangement of the TIS11B granules and the ER enables the ER to act as diffusion boundary for SET. This traps SET at the ER surface and facilitates 3'UTR-mediated protein-protein interactions between SET and newly made membrane proteins, including CD47-LU and PD-L1.

The function of TIS11B in the destabilization of mRNAs does not require TIS11B granule formation. Therefore, when RNA-binding proteins are soluble or present as single entities, they can regulate mRNA-based processes, including mRNA stability. However, assembly of RNA-binding proteins into larger aggregates, including RNA granules, allows them to acquire new properties that are necessary for the regulation of protein functions through 3'UTR-dependent protein-protein interactions.