A recent study by scientists from the International Centre for Genetic Engineering and Biotechnology (ICGEB) published in the journal Nature Scientific Reports sheds light on the mechanisms employed by a family of proteins, called FET proteins, to recognise and bind to RNA. 

FET proteins have amino acid sequences that are preserved across species. These proteins, which play an important role in the normal housekeeping of the cell, are implicated in various oncogenic diseases. Mutations in some members of this family — FUS (Fused in sarcoma) or TAF15 (Human TATA binding protein associated factor 2N) — are also known to be involved in neurodegenerative disorders like amyotrophic lateral sclerosis (ALS) and frontotemporal lobar dementia. FET protein functions range from transcribing DNA to processing RNA sequences with roles such as translation, splicing, transport, stability and localisation. Little was known, at a molecular level, about how these proteins recognise their RNA targets and bind to them.

In macromolecular interactions, like protein-protein or protein-RNA complexes, the interactive forces could be weak, like Van der Waals, electrostatic, hydrophobic or ionic. Strong bonds arise from covalent interactions, such as disulphide bonds. Most RNA binding proteins have multiple binding sites, called domains. To aid binding, the RNA binding domains (RBDs) carry protruding amino acid residues which facilitate weak or strong interactive forces. For most of the RBDs, these interactions are known to be RNA sequence specific. 

Researchers in this study used a suite of tools — solution-state NMR spectroscopy, calorimetry, docking and molecular dynamics simulation — to delineate specificities of RBD-RNA interactions in the FET protein TAF15. What they found was that in a manner unique to TAF15, the predominant RBD in this protein, known as RRM (RNA recognition motif), forms hydrogen bond (H‑bond) pairs with RNA. The RNA part in this interaction are stem loops — hairpin-shaped intramolecular base pairing patterns — known as the SON (stem in natural left-right configuration of RNA). The bindings were based on RNA’s secondary structural configuration and independent of protein backbone. When further analysed using different RNA sequences, the interactions were found to be independent of the length of the RNA sequence as well. This mechanism of RNA binding is distinctive — other RNA binding proteins are known to rely on non covalent interactions called pi-pi stacking. Since the RRM domains are highly similar in TAF15 and FUS, the authors believe the mechanisms of protein-RNA binding in FUS to be similar as well.

Senior author and Staff Research Scientist, Neel Sarovar Bhavesh described this mechanism of binding with an insightful analogy: ​“It would be like holding a cricket ball in a glove. TAF15 binds tightly or clamps to the stem loop of RNA, stabilises it and regulates protein production. This interaction is facilitated by the concave face, or the glove, of the protein. The mode of interaction is thus a form of ​‘concave-convex’ lock and key model.”

How does this increase our understanding of the neurodegenerative disorders associated with this family of proteins? Bhavesh explains, ​“Point mutations in RRM surface are known to be implicated in ALS. It is possible that with these mutations, the concave surface of the RRM domain in TAF15/FUS doesn’t form stable interactions with RNA and other protein complexes. These destabilising binding interactions could possibly lead to aggregation of RNA/​protein complexes that are a hallmark of ALS.”

While the study did a full structural analysis of TAF15 and eliminated roles of other domains in RNA recognition and binding, the group hopes to work out the full mechanistic details of FET proteins at the cellular level as well. Using different RNA sequences and point mutations, details on what stages of the interaction regulation starts, or at what stages the protein aggregation might occur, would help in understanding of progression of associated neurodegenerative diseases.