While our genome is pervasively transcribed into tens of thousands of non-coding RNAs, only a few of these transcripts have so far been characterized at the molecular level. For instance, only 1% of all available high-resolution 3D structures in the PDB are RNA structures. As a result, we currently still have a very poor understanding of RNA structure, folding, small molecule recognition, and functional mechanisms.

RNA presents unique experimental challenges that cannot yet be satisfactorily addressed by bioinformatics and computational tools, as it is the case for proteins. For instance, it is difficult to predict RNA evolutionary conservation, to compute RNA dynamics, and to model RNA secondary and tertiary structures accurately, despite advances in artificial intelligence algorithms. Yet, RNA-based therapies are becoming a reality opening unprecedented frontiers in personalized human medicine. So, RNA needs to be considered as a strategic subject for experimental research and we should fill the gap in structural biology that separates RNA from proteins.

To fill this knowledge gap, our laboratory pursues two main research lines.

1. Characterizing and modulating the fundamental process of splicing, in prokaryotes and eukaryotes

Here, we characterize the self-splicing group II introns. These ribozymes are bacterial and organellar ancestors of the nuclear spliceosome and retro-transposable elements of pharmacological and biotechnological importance. Integrating enzymatic, structural, and simulation studies, we have shown at the molecular level how these 400-nt-long ribozymes fold, perform catalysis, and recognize small molecules through their conserved active site. Our work provides an unprecedented molecular movie of these multidomain RNAs assembling into functionally active conformations and establishes a basis for understanding how RNA avoids the formation of non-functional ‘kinetic traps’ during folding. Furthermore, we have resolved five different stages of catalysis, thus visualizing the atomic details of the splicing mechanism in bacteria and organelles. Finally, we have elucidated the molecular basis for bacterial and organellar splicing inhibition by small RNA-binding molecules, which acts selectively and specifically at the conserved catalytic site by adopting distinctive poses at different stages of catalysis. Our insights into the folding, mechanism of splicing, and mechanism of splicing inhibition are in line with biochemical, structural and functional data on the spliceosome. The correspondence between group II introns and spliceosome data reinforces the notion that these evolutionarily-related molecular machines share the same enzymatic strategy and can potentially be modulated by small molecules of the same chemical class. Our work thus provides a solid basis for the rational design of splicing modulators not only against bacterial and organellar introns, but also against the human spliceosome, which is a validated drug target for the treatment of congenital diseases and cancers.

Using toolkits from the characterization of splicing ribozymes, we also aim to obtain high-resolution structures of long non-coding RNA (lncRNAs) because such an achievement would enable us to understand and modulate lncRNA mechanisms of actions at unprecedented details, including at the molecular and atomic levels (see research line 2).

2. Characterizing the molecular mechanism of action of human lncRNAs.

My group is studying the molecular mechanism of mammalian lncRNAs involved in gene expression regulation during development and in response to environmental stress. We are particularly interested in how lncRNAs fold into discrete secondary and tertiary structures and how these structures determine cellular functions in association with regulatory proteins, such as chromatin remodeling enzymes and transcription factors.

While we primarily use biochemical approaches, such as SHAPE and HRF probing, in combination with state-of-the-art biophysical methods, such as X-ray crystallography, cryo-electron microscopy, small angle scattering, atomic force microscopy, and NMR, we also crucially integrate these in vitro studies with in vivo functional assays, such as gene expression and cell cycle analysis.

In a recent work, we have determined novel and unexpected secondary and tertiary structure properties of an imprinted alternatively-spliced human lncRNA named MEG3 (maternally expressed gene 3). Thanks to a systematic mutagenesis approach coupled to high-throughput in vivo functional assays and evolutionary alignments, we have surprisingly discovered that MEG3 needs to adopt a discrete tertiary structure to be able to stimulate protein p53, thus acting as a tumor suppressor in adult cells. In parallel, we are producing known interactors of MEG3, such as Polycomb group proteins and p53 itself, to characterize these proteins and their interactions with MEG3 from a biochemical and biophysical point of view.

Our achievements now open the way for characterizing the structure and function of MEG3 and of many other lncRNAs at greater level of precision, by high resolution biophysical techniques and proteomics, transcriptomics, and genomics approaches. Our aim is to decipher at least some of the puzzles and mysteries surrounding the biology of lncRNAs. What are the molecular determinants for the apparently promiscuous, but tight and selective interaction of lncRNAs with their protein cofactors? What controls lncRNA folding in the crowded nuclear environment? What role does the complex structure of lncRNAs play in orchestrating the interplay of all epigenetic regulators and the structure of chromatin? Ultimately, obtaining high resolution insights into lncRNAs and their ribonucleoprotein complexes will allow the measuring of epigenetic processes quantitatively and will potentially lead to the development of new therapeutic approaches to cure invasive human diseases.