RNA-binding proteins (RBPs) are among the key regulators of post-transcriptional processes and help cells to adapt to changing internal and external stimuli. The exploration of how RBPs perform their complex tasks of organising alternative splicing, polyadenylation, mRNA localisation, translation and degradation is crucial, as they are fundamental gene- regulatory processes. To unravel the function of a specific RBP, it is necessary to determine its binding sites and its regulatory effects on target transcripts. So far, many techniques have been protein-centric and required the immunoprecipitation of an individual protein with its interacting RNAs (CLIP-based techniques). However, approaches based on the principle that individual RNAs are only bound by one RBP at any point in time are too simplistic. Most RNAs are bound by a variety of different RBPs at the same time or sequentially throughout their life cycle. Therefore, it is essential to identify the pool of all RBPs that directly bind to a specific RNA in order to understand the RNA’s regulation. So far, techniques used RNA as bait in order to identify interacting proteins. The RNA is immobilised on a solid support and after thorough washing, the RNP complexes (RBPs bound to RNA) can be identified through mass spectrometry. Alternatively, the RNA can be tagged with e.g. modified ribonucleosides. Even though these techniques are able to identify in vivo and in vitro RNA-RBP interactions, they suffer from several disadvantages, such as perturbations of RNA folding through the insertion of a tag, the need for large amounts of input material, limited identification of non-abundant targets, and limited throughput. To overcome such limitations, I proposed to develop a new method, which aims at uncovering all of the RBPs bound to a specific RNA without the need for RNA or protein enrichment. This novel technique should allow the unbiased identification of RBPs directly bound to a specific RNA target within cells.
In 2014, Dr Mikhail Savitski, group leader at EMBL, developed a novel approach, termed thermal proteome profiling (TPP). This methodology is based on the principle that the binding of a ligand to a protein causes a shift in its stability, which can be read out by the proteins’ denaturing properties, using the cellular thermal shift assay (CETSA). In detail, target engagement is measured by exposing the intact cells to a range of different temperatures (37¬¬–65°C, at increments of 3°C). Thereby, a specific temperature induces unfolding of a protein, which leads to the formation of an insoluble pellet after centrifugation. The supernatant now contains all the proteins that are stable at the given temperature, which can be quantified either by Western blotting for specific targets, or on a proteome-wide scale using mass spectrometry. The addition of a drug, as in Savitski et al., (2014), perturbs the system and results in specific protein melting point differences.
The project proposed here made use of this innovative approach, by transferring this cutting-edge technology to the study of RNA-protein complexes and identify all of the RBPs that are bound to a specific RNA. One major advantage of this application is that the differential effects on protein stability modulated by the concentration of a specific RNA molecule can be measured in the context of the whole cell. Specifically, we optimised the technique with a well-characterized model system, the iron regulatory protein/iron response element interaction, followed by the application of TPP to discover all RBPs bound to the iron response element of the HIF2 5’untranslated region. Trying to apply TPP to a novel context, directed our interest to Enolase 1, a glycolytic enzyme that binds RNA.
