Legionella pneumophila secretes ~300 effector proteins into the host cytosol during infection. These effectors exert multitude of biochemical activities and drive the maturation of Legionella containing vacuole (LCV) by hijacking several host defense pathways. Deletion of the effector SidJ negatively impacts legionella’s intra cellular growth. We recently showed that SidJ acts as a glutamylating enzyme that adopts a novel fold with distant similarity to kinase domains. SidJ shows no sequence or structural homology to mammalian glutamylating enzymes and contains two nucleotide-binding pockets. My proposal aims to address how SidJ recognizes its target proteins for glutamylation, how the two ATP-bindings pockets of SidJ co-ordinate to achieve target protein glutamylation and what host proteins does SidJ target especially after early stages of legionella infection.

The germline of a wide variety of animals express a special class of small RNAs called piwi-interacting RNAs (piRNAs). Its presence in the murine germline was shown in early 2006, and since then, intense research activity is focused on understanding the role of these RNAs. Initial studies show that they fundamentally differ in their biogenesis and function from the well-known microRNAs (miRNAs) and small interfering RNAs (siRNAs). It is not clear how murine piRNAs are made and the factors involved have not been identified. One of the roles attributed to piRNAs is the maintenance of genome integrity by silencing of transposons and other repeat elements by promoting DNA methylation of these elements. We have set out to biochemically characterize Piwi-associated proteins as a starting point in understanding the mechanism by which piRNAs act in this pathway. In one such complex we identified a tudor domain containing protein called Tdrd1; a mouse mutant of tdrd1 shows loss of transposon silencing and concomitant absence of DNA methylation. This is the first such factor (apart from Piwi proteins) identified that directly links it to piRNA function. Another factor identified is an RNA helicase, the homolog of which is already known to participate in the miRNA pathway. Thus, these two findings validate our biochemical method as a very useful approach in studying the piRNA pathway. We have also generated mouse mutants that will yield insight into the in vivo roles of this small RNA pathway in the germline. As there is no cell culture system available for germline studies, we are generating a GFP-based reporter for studying piRNA function in zebrafish. This required us to set up a new fish lab in Grenoble and learn to handle fish. All these are described as preliminary data and some studies are submitted for publication. In summary, this proposal aims to use interdisciplinary approaches covering biochemistry, computational methods, and genetics (mouse and zebrafish) in achieving its goals.

The proper functioning of any living cell requires that approximately half of all proteins synthesized in the cytosol must be translocated across (in the case of exported proteins) or inserted into (in the case of membrane proteins) a cell membrane. Protein translocation at the membrane occurs through a proteinaceous channel, termed the translocon. The core of the translocon is a heterotrimeric integral membrane protein complex (SecY, SecE and SecG in eubacteria) and is conserved in all kingdoms of life. Little is known about the structure and function of the additional components of the holo-translocation machinery, SecD, SecF and YidC, which are essential for eubacteria. In spite of their central role, the function of these subunits is entirely unresolved to date. This is to a large part due to the lack of a purified, reconstituted SecYEG-DF-YidC holo-translocon, and the absence of any structural data on this large transmembrane multiprotein complex. Using a new recombineering-based vector system for expression of multi-protein complexes in E. coli, we now for the first time successfully over-expressed and purified the SecYEG-DF-YidC holo-translocon. With this purified holo-translocon in hand, we are now in a unique position to carry out a thorough functional and structural characterization of this vitally important membrane protein complex. Our project will be divided in five complementary parts: (1) We will solve the structure of the holo-translocon by cryo-electron microscopy and single particle analysis. (2) We will study co-translational translocation by solving the structure of the holo-translocon complex with a translating ribosome by cryo-electron microscopy and single particle analysis. Using the existing structures of the ribosome, SecYEG and the predicted homology of SecDF to AcrB, we will build a quasi-atomic model of the holo-translocon. (3) We will thoroughly assay the molecular level function of the holo-translocon in protein translocation, in a collaborative setup with Prof. Ian Collinson, Bristol, UK. The functional modulation of SecA-driven post-translational translocation by SecDF and the effect of the proton motive force will be investigated. (4) The stoichiometry of the subunits in the holo-translocon complex is an entirely unresolved issue to date. We will analyze the stoichiometry and the architecture of the holo-translocon by mass spectrometry, in collaboration with Prof. Carol V. Robinson, Cambridge, UK. (5) We will perform crystallization experiments of the holo-translocon for analysis by 2D electron crystallography (with Prof. Ian Collinson, Bristol) and for 3D X-ray crystallography by using the high-throughput crystallization facilities that are part of the Partnership in Structural Biology (PSB) here in Grenoble. A molecular understanding of protein export and membrane protein integration and folding will not only provide fundamental insight into a paramountly important biological process, but may also contribute significantly to the future design of novel, and urgently needed antibiotics.

Trypanosomiasis causing parasites in humans and farm animals represent a severe socio-economical threat in the southern hemisphere and are difficult to treat with currently available drugs. On a cellular level, certain RNA processing pathways differ significantly between humans and the parasite – a vulnerable Achille’s heel that could be targeted by new therapeutics for treatment. Accumulating cell biology data demonstrates the importance of these pathways, however, atomic structures of the complexes involved are missing. Here, we aim to structurally, using X-ray crystallography and cryo electron microscopy, and biochemically characterise two essential trypanosomal RNA processing machineries that are unique to the parasite: RESC, a complex required for mitochondrial mRNA editing and NCBC, a complex vital for trans-splicing. Beyond gaining atomic resolution detail into their mechanism of action, in the future, we will target them by structure-aided design of new anti-parasitic drugs.

Arenaviridae, Bunyaviridae and Orthomyxoviridae are the principal families of segmented negative strand single-stranded RNA viruses and include many serious and/or emerging human pathogens (e.g. respectively Lassa Fever, Crimean-Congo Hemorrhagic Fever and Influenza viruses). For arena- and bunyaviruses, transcription and replication of the viral genome segments is performed by the single chain viral RNA-dependent RNA polymerase or L-protein, which is functionally similar to the hetero-trimeric influenza polymerase complex. Two distinctive features of all these polymerases is that they bind to the conserved, quasi-complementary, non-translated, 3’ and 5’ ends of each vRNA segment and that they perform transcription by ‘cap-snatching’, unlike the polymerases of non-segmented negative strand RNA viruses which possess 5’ capping activity. Despite years of study, particularly on influenza polymerase, a detailed understanding of the mechanisms of transcription and replication is lacking, largely due to the absence of high resolution structural information. In recent years, several structures have been solved of functional domains of influenza polymerase whereas the L-proteins of arena- and bunyaviruses are almost ‘virgin territory’, apart from the cap-snatching endonuclease which has recently been identified by the partners of this proposal to be at the extreme N-terminus of L and is very similar to that of influenza polymerase. This discovery has reinforced the hypothesis that the polymerase of all segmented, negative strand, single-stranded RNA viruses probably have a similar architecture. To explore this hypothesis further we plan to apply a battery of state-of the-art structural biology techniques, to derive structural information (notably by X-ray and electron microscopy) on arena- and bunyavirus L-protein at the level of functional domains as well as that of the full-length L-protein and its complexes with vRNA. Mutagenesis based on the structural results will be used to drive in vitro biochemical and in cell viral replication studies to test functional hypotheses. The long-term, ambitious goal of this project is to derive a structure-based, mechanistic model to explain how segmented, negative-strand RNA viral RdRps, in the context of complexes with nucleocapsid (N) and vRNA, function in both transcription by cap-snatching (giving rise to translation competent viral mRNAs) and replication (giving rise to full-length copies of the genome) and how the switch between the two modes of operation is mechanistically achieved and regulated. Preliminary results on domain identification and expression, purification and EM of full-length bunyavirus L-protein suggest that this system may prove more tractable than influenza virus polymerase. Furthermore it is part of the strategy of the project that each partner focuses on one of the two evolutionary diverged, but related, Arena- or Bunyavirus systems, thus increasing the chance of success as well as giving the opportunity of cross-fertilisation from one system to the other. Finally, the structural information already available on the cap-snatching endonuclease of arena and bunyaviruses, and possibly on other functional domains, will be used to optimize inhibitors as a first step towards antiviral drug design, using as a starting point the several families of inhibitors that are known to target the similar influenza virus endonuclease. In this respect it is important to remember that Arena- and Bunyavirus are largely rodent, bat or insect borne and in a world of environmental change, newly emerging threats from these families of viruses might be expected.