Replication of many viruses occurs in specialized compartments formed during infection and known as viral factories. The physicochemical nature of these factories and the molecular basis of their morphogenesis and organization are poorly understood. The Mononegavirales (MNV) order includes several important human pathogens (Rabies virus -RABV-, Measles Virus -MeV-, Ebola virus…). All these viruses have a single strand RNA genome of negative polarity which is encapsidated by the nucleoprotein (N) to form the ribonucleoprotein that is associated with the RNA dependent RNA polymerase and its cofactor the phosphoprotein (P). RABV factories are the Negri bodies (NBs) which are cytoplasmic inclusions housing the synthesis of viral RNA (mRNAs and genomic RNAs). We have demonstrated that NBs constitute a new category of membrane-less liquid organelles. Liquid organelles are formed by liquid-liquid phase separation (LLPS) and contribute to the cell compartmentalization. They are involved in a wide range of cellular processes. So far, the general principles leading to LLPS are poorly understood. Published experimental data indicate that the liquid nature of viral factories can be generalized to other MNVs. This is a paradigm shift which opens new research horizons in the field of MNV replication and invites us to revisit the interplay between viral factories and the components of the cellular innate immunity. This proposal aims to characterize the morphogenesis, the internal organization, the composition, the dynamics and the functions of RABV and MeV viral factories. It brings together 3 teams: a team of virologists and biochemists specialized in rhabdoviruses, another which develops new methods for cell biology on synchrotron radiation sources and a third one developing state-of-the-art solution state NMR and fluorescence approaches to investigate the dynamics and interactions of highly flexible proteins with a strong focus on negative strand RNA viruses structure. The project has six major objectives: 1) Using super-resolution microscopy and focused ion beam scanning electron microscopy, we will characterize the submicrometer organization of NBs and how it evolves all along the viral cycle. NBs ultrastructure will also be investigated by various methods developed on synchrotron radiation sources (e.g. scanning transmission X-ray microscopy or µ-SAXS coupled to correlative imaging). We will also determine viral proteins structural elements which are required for NBs formation. 2) Using in vitro reconstituted systems and a combination of fluorescence techniques and NMR, we will identify the physicochemical principles underlying the LLPS leading to the formation of the viral factory. 3) As LLPS enriches NBs in specific factors, we will characterize NBs’ proteome using a proximity biotinylation assay and identification of proteins by mass spectrometry. We will also identify RNAs which are NBs’ residents. 4) We will characterize the interplay between NBs and cellular innate immunity. Indeed, the sequestration of viral RNAs in NBs raises the question of their accessibility to pathogen recognition receptors such as RIG-I and MDA5. Alternatively, liquid factories might constitute a signature of viral infection and cells might have evolved a mechanism allowing the sensing of such structures and/or their destabilization. 5) Experimental data suggest that RABV factories might also harbor viral proteins translation. We will investigate where viral mRNAs are translated in infected cells and if they use an unconventional mechanism of translation initiation to escape the translation inhibition induced by innate immunity. 5) At some stage, the RNPs must leave the viral factory to form new virions. We will investigate the molecular bases of the processes by which this happens. Beyond its impact in virology and innate immunity, this project should have an impact in cellular biology by increasing our understanding of liquid organelles assembly.

Flowers ensure angiosperm reproduction and are the basis for fruits and seeds. The LEAFY transcription factor orchestrates flower development in angiosperms. Its activity highly depends on its physical interaction with the UFO ubiquitin ligase but the reason for this has remained obscure. Our project aims at understanding how LFY and ubiquitination pathways interact to make flowers. We will use a combination of genomics, imaging, structural biology and proteomics to understand i) where this interaction occurs at the tissue, subcellular and genomics levels ii) the biochemical and structural properties of the UFO-LFY complex iii) how it controls gene expression iv) the function of ubiquitination. Moreover, we propose to create a tool combining LFY and UFO able to trigger the development of flowers from any tissue allowing to modify inflorescence structures at will, bypassing species or environmental constraints.

The asymmetric distribution of lipids between the two leaflets of cell membranes is a fundamental feature of eukaryotic cells. For instance, while phosphatidylcholine and sphingomyelin are restricted to the outer leaflet of membranes of the late secretory/endocytic pathways in most cell types, phosphatidylserine (PS), phosphatidylethanolamine, and phosphatidylinositol-4,5-bisphosphate are only found in the cytosolic leaflet. Regulated exposure of PS in the outer leaflet of the plasma membrane is an early signal for clearance of apoptotic cells by macrophages or triggering of the blood coagulation cascade. Inside the cell, PS plays critical roles since the high negative surface charge conferred by PS on the cytosolic leaflet of membranes facilitates the recruitment of polybasic motif-containing proteins such as the small GTPase K-Ras and the membrane fission protein EHD1, providing a link between PS distribution and regulation of cell signalling and vesicular trafficking. For transbilayer lipid asymmetry to be maintained, cells have evolved the so-called lipid flippases, transmembrane proteins from the P4-ATPase family which are responsible for the active transport of lipid species from the exoplasmic to the cytosolic leaflet of membranes, at the expense of ATP. Most P4-ATPases require association with transmembrane proteins from the Cdc50 family for proper localization and lipid transport activity. The yeast lipid flippase complex Drs2-Cdc50 has been shown to specifically transport PS and this transport is crucial for bidirectional vesicle trafficking between the endosomal system and the trans-Golgi network (TGN). Mutations in human P4-ATPases have been linked to severe neurological disorders, reproductive dysfunction as well as metabolic and liver disease, underlining the essential role of transbilayer lipid asymmetry in cell physiology. We previously showed, using a combination of limited proteolysis, genetic truncation, and structural approaches, that the catalytic activity of purified Drs2-Cdc50 complex is autoinhibited by its two unstructured N- and C-terminal extensions and activated by phosphatidylinositol-4-phosphate (PI4P). Yet, the molecular mechanism underlying activation of Drs2-Cdc50-dependent lipid transport activity remains unknown. Recently, the small GTPase Arl1 and the Arf-GEF Gea2, a GDP/GTP exchange factor for Arf, were shown to physically interact with the N- and C-termini of Drs2, respectively, and to be required for Drs2-Cdc50-catalyzed lipid transport in isolated TGN vesicles. Arl1 also binds to Gea2, suggesting an intricate mechanism for the regulation of Drs2-mediated transbilayer lipid transport. Based on previous work and our preliminary results, our working hypothesis is that binding of Arl1 and Gea2 to the N- and C-termini of Drs2 relieves autoinhibition and thus activates lipid transport by Drs2-Cdc50. Hence, combining biochemical, in silico and medium/high-resolution structural approaches, FLIPPER aims to dissect this regulatory mechanism, using in vitro reconstitution of the lipid transport machinery. This will be achieved by combining our expertise in the structural and biochemical analysis of small GTPases and Arf-GEFs (J. Cherfils) with structural mass spectrometry techniques, including hydrogen-deuterium exchange mass spectrometry (C. Bechara), structure determination of the Drs2-Cdc50-Arl1-Gea2 complex by cryo-EM (J. Lyons/P. Nissen) and know-how into the biochemistry and functional investigation of lipid flippases (G. Lenoir). Altogether, our proposal aims to provide a mechanistic basis for Drs2 activation in vivo and reveal new functions for understudied small GTPases and large Arf-GEFs such as Arl1 and Gea2.

Mammalian genomes are divided into Topologically Associated Domains (TADs) that restrict the formation of Enhancer-Promoter loops and other biological processes. TADs are formed by a process of Cohesin-mediated loop extrusion, which is subsequently blocked at defined TAD boundaries. Most TAD boundaries bind the CTCF insulator protein, but features like G-quadruplexes (G4s) and transcriptional start sites are enriched as well. We recently reported that CTCF binding is clustered at most TAD boundaries, both at the level of ChIP-seq peaks and of DNA binding motifs within peaks. Clustering of binding motifs within peaks was particularly beneficial for CTCF binding at lower affinity motifs, suggesting that this local clustering may create binding synergies. Using Nano-C, a new multi-contact 3C assay, we showed that TAD boundaries are not impermeable and that CTCF binding peaks contributed individually (but incompletely) to the blocking of loop extrusion. Clustering of peaks thereby improved the overall capacity for loop extrusion blocking, thereby enhancing the separation between TADs. We hypothesize that loop extrusion blocking at TAD boundaries can be defined by their grammar of CTCF binding (‘Insulator Grammar'), with chromatin features like nucleosome positioning, transcription and G4s having a further impact. This DNA-encoded grammar provides a flexible means for the regulation of TAD boundary permeability, thereby allowing the fine-tuning of enhancer-promoter loop formation and gene regulation. Our InsulatorGrammar project aims to systematically dissect how the CTCF grammar can be used to regulate Cohesin-mediated loop extrusion, thereby allowing the modulation of enhancer-promoter loop formation. To address this aim, we will determine how the DNA-encoded local clustering of CTCF binding motifs, in combination with other chromatin features, creates favorable chromatin environments for CTCF binding, and how this synergy affects loop extrusion and enhancer blocking. First, we will first use cells that permit a timed degradation and reestablishment of the CTCF protein to determine, genome-wide, how the CTCF binding grammar (number, orientation and affinity of motifs) and other chromatin features reciprocally influence CTCF binding. Next, we will use a newly developed reporter system for enhancer-promoter looping to integrate 50 synthetic ‘designer’ insulator elements that vary for the number, orientation and affinity of CTCF motifs, and in the presence or absence of nearby transcription and G4s. For all lines, the influence of the CTCF grammar on CTCF binding and enhancer blocking will be determined. Moreover, one-third of these lines will be analyzed using a strategy that combines in-depth multi-omics with biophysical modeling to unravel underlying characteristics of chromatin organization. These studies will reveal how the CTCF grammar and the surrounding chromatin features reciprocally affect the DNA binding of CTCF, the positioning of nucleosomes and the formation of G4s. These structural insights will subsequently be linked to the functional impact on loop extrusion blocking, enhancer-promoter looping and TAD organization. Our multi-disciplinary InsulatorGrammar project will generate an unprecedented view of how clustering of CTCF binding sites can synergize with other chromatin features to create a DNA-encoded means for the fine-tuning of enhancer-promoter looping and gene regulation. The results will help to explain how more subtle changes in gene activity can be introduced, with relevance for development, evolution and disease.

Many pathogens manipulate the cAMP intracellular signaling of host cells to promote their survival and proliferation in hosts. Bacterial ExoY-like virulence factors represent a new atypical subfamily of nucleotidyl cyclase (NC) toxins. Exoenzyme Y (ExoY) was first identified as a toxin secreted via a type 3 secretion system (T3SS) by Pseudomonas aeruginosa, a major opportunistic and nosocomial human pathogen. P. aeruginosa is also responsible for progressive and severe chronic lung infections in patients with cystic fibrosis. ExoY is present in 90-98% of clinical isolates of P. aeruginosa, which suggests an important role in bacterial pathogenicity, but its role in P. aeruginosa infections remains poorly understood. Poorly characterized ExoY-related enzymatic modules or effector proteins have also been found recently in several toxins produced by various Gram-negative bacteria representing emerging human or animal pathogens. Some of these enzymes were shown to be essential for bacterial virulence. We have recently shown that these NC toxins are potently activated within the host target cells by using an original eukaryotic cofactor that is actin. Yet, our new preliminary data suggest that they may differ significantly in their substrate selectivity, their interaction with actin and activation mechanisms, impact on the actin cytoskeleton dynamics, and subcellular localizations. As a consequence, they likely perturb different biological processes in infected host cells. We aim here to characterize at the molecular and structural level and in cellular infection models the functional specificities, precise role and virulence mechanisms of this novel class of actin-activated NC toxins in bacterial infections by P. aeruginosa and various pathogenic Gram-negative organisms. To decipher their structure-function relationships we will analyze several representative members of this subfamily of actin-activated NC toxins. Our main goals are to characterize in vitro and in cells their functional specificities and impact on actin self-assembly dynamics. We will study the structural bases for their activation mechanisms and enzymatic specificities, and search for inhibitors. We will characterize ExoY phenotype and prevalence in the strains included in the P. aeruginosa reference panel collection. We will study ExoY effects on phagocytosis, inflammation, mucus production, or the integrity of cell-cell junctions in P. aeruginosa-infected cells to determine the importance of ExoY in the modulation of host airways innate immunity. We will finally analyze possible interplays between ExoY/ExoY-like virulence factors and other bacterial virulence factors that affect the actin cytoskeleton. Collectively, our proposed research will be innovative in several aspects: it will expand the fundamental knowledge on Pseudomonas aeruginosa T3SS effector virulence mechanisms in P. aeruginosa infections; it will help to identify the functional specificities of ExoY-related toxins recently found in emergent human or animal proteobacterial pathogens, and will eventually explore their potential as drug target. Finally, our project may help to understand better the pathogenicity of P. aeruginosa in diseases such as cystic fibrosis for which there is yet no effective therapy.