The association between the development of metastases and the risk of thrombotic complications has been documented a long time ago. Armand Trousseau was the first in 1865 to establish a direct correlation between thrombophlebitis and development of cancers. Since that time, the close relationship linking cancer, metastasis and thrombosis has been the focus of several studies. Kaplan et al. have recently developed a new concept named formation of 'premetastatic niches' to describe how a cancer cell will leave its primary tumor location to reach a secondary site where it will form a metastasis. The authors proposed that cancer cells secrete pro-metastatic factors that were found in plasma of tumor bearing animals. Recent clinical publications have demonstrated the presence in blood circulation of patients suffering from cancer of tissue factor bearing microparticles. The authors suggest that these microparticles are directly responsible for the development of a procoagulant state. However, the origin of these microparticles and their direct involvement in thrombosis and metastasis processes are to date unknown. We hypothetise that tissue factor bearing microparticles produced by cancer cells play a key role in both formations of a thrombotic state and of pre-metastatic niches. To solve that problematic we will performed in vitro experiments, used animal models and state of the art techniques of digital real time intravital microcopy that we recently import from the US to our current laboratory. Our first results, currently in revision in JEM, show that mice show that human and mice pancreatic cancer cells produce microparticles expressing tissue factor and PSGL-1. In vivo, cancer cell-derived microparticles, but not their parental cells, infused into a living mouse accumulate at the site of injury and reduce both the tail bleeding time and the time to occlusion of veins and arteries. This thrombotic state is also observed in mice developing a tumor. In such mice, cancer cell-derived microparticles shed from the growing tumor are able to enter the circulation and accumulate at the site of injury. Infusion of a blocking P-selectin antibody inhibits the interaction of cancer cell-derived microparticles with platelets and abolishes the thrombotic state observed both after injection of microparticles or when mice developed a tumor. We will now continue to focus on a mouse model of tumor induced by different pancreatic neoplasic cells. One of our first aims will be to continue to identify in vitro proteins, such as PSGL-1, potentially involved in thrombosis and/or adhesion processes and expressed on the surface of neoplasic cells derived microparticles. These proteins may play a critical role in activation/aggregation of platelets and formation of the premetastatic niches by interacting with (activated) endothelial cells. Based on our preliminary results and on current scientific literature we will focus on tissue factor (TF) but also, its main inhibitor, tissue factor pathway inhibitor (TFPI) and its recently described regulator, the protein disulfide isomerase, PDI. The presence and role(s) of the oncoisoforrm of the Bile salt dependent lipase (BSDL) will also be studied. Indeed, we reported, few months ago, that BSDL could interact via its V3 like loop structure with the chemokine receptor CXCR4, receptor known to play an important role in development of metastasis. Last, the roles of adhesive proteins (Fibronectin, integrin '2'1, P-selectin/PSGL1) or coagulant proteins (Urokinase Pathway Activator (UPA) and its receptor, UPAR) will be determined. Parallel to this work we will used our intravital microscope to set up mice models in which, endogenous and exogenous microparticles derived cancer cells, endogenous or exogenous cancer cells, endogenous platelets and endothelium will all be specifically detected and image in vivo. The main goal of this part is to confirm our initial observation that cancer derived microparticles are involved in thrombosis and metastasis processes by following and imaging in real time in vivo interactions between the different partners. Last, once we have identified proteins expressed on microparticles and set up our models we will confirm, using siRNA or blocking antibodies, in vivo the role played by these proteins and will dissect molecular mechanisms involved in thrombosis and metastasis. To perform this project we will actively used the expertise of people present in our current laboratory (INSERM UMR 911) that focus on pancreatic cancer since more than a decade. We will also use the resources (intravital microscopy, electron microscopy'), and the world recognized expertise of people present on site (collaboration with Prof. Francoise Dignat-George). Our ultimate goal is to identify new target(s) of clinical interest to limit both the development of a thrombotic state and the formation of metastasis.

BACKGROUND Spinal cord injury (SCI) is not a common condition, but it has a major effect on the injured person's functional, medical, financial, and psychosocial well-being. Spinal cord injury is an insult to the spinal cord that can occur through different mechanisms, including contusion, compression and ischemia. SCI causes either temporary or permanent damage to white myelinated fiber tracts that carry sensation and motor signals to and from the brain; it also damages vasculature, as well as gray matter in the central part, causing segmental losses of inter and motor neurons. The traumatic insult leads to functional disruptions below the site of injury such as loss of voluntary movements and muscle atrophy, as well as functional deficits such as cardiovascular system, breathing or pain control. Within weeks or months, endogenous repair (by neuronal recovery and neovasculature) as well as functional recovery may occur and gradually returns to control levels. The last decade has seen many advances in understanding, visualizing and treating spinal cord injuries, however a better description and characterization of the pathology is still necessary. OBJECTIVE The general objective of this project is to develop a non invasive in vivo spinal cord (SC) multimodal MR imaging strategy, which will provide a better characterization of the tissue alterations and repairs that occur after the injury. By taken advantage of the non-invasiveness and the multi-parametric investigation properties of MRI, the proposed project would present, in case of successfulness, an alternative approach to the techniques usually used for the investigation of SCI animal models. Despite their high sensitivity and being well-adapted to inherent SC challenges (size, complex structure...), most of the used techniques (histopathology, optical imaging') suffered from main limitations including invasiveness properties and limitative field of investigation. Therefore, the ability of performing longitudinal studies, although crucial in SCI investigation, is limited, so is the variety of physiological investigated parameters. The proposed project and the associated MRI technique might give some solutions to the previously described limitations. This project appears then very novel and ambitious. Important fundamental advances are expected, in particular on animal models, but also on clinical studies, MRI protocols offering the advantage of being directly adapted for human studies. DESCRIPTION OF THE PROJECT The proposed project will be conducted according to three main axes. The first step will concern the development of a multimodal MRI approach for the pathological mouse SC exploration. Based on classical high-resolution MRI techniques (anatomy, relaxometry and diffusion imaging), the project will also integrate 2 ambitious techniques: perfusion imaging and NMR spectroscopy. The overall applied MR techniques will provide non-invasively and in vivo, morphological, structural, metabolic and hemodynamic parameters relative to the physiological state of the spinal cord. The second step will focus on the establishment of correlations between the evolution of MR parameters, with functional evolution and histopathology studies, in the case of SCI compression model. The sensitivity of the MR techniques will be a key point for the success of this part since characterization of mild lesions, detection of early changes and fine discrimination between 3 groups of mice are expected: non-treated mice, naturally trained mice placed in an enriched environment and pharmacologically treated mice. The correlations that should be observed between the SC lesion deficit/repair (identified by MR SC investigations), the motor, muscular and sensory alteration/recovery (observed by functional test) and the histological evaluation will furthermore permit to point out on the most relevant MR parameters involved in the regeneration processes. Finally, the third step will be the transfer and development of similar multimodal MR methods on the clinical MR scanners for investigation of human spinal cord injury.

Salmonellosis is one of the most common and widely distributed bacterial foodborne diseases. Since the beginning of the 1990s problems related to Salmonella have increased significantly and new concerns have been identified. Strains of Salmonella which are resistant to a range of antimicrobials, including first-choice agents for the treatment of humans, have emerged and are threatening to become a serious public health problem. To face this problem much research effort has been made in recent years towards understanding Salmonella pathogenesis and especially the interplay between this pathogen and host cells, expecting it will lead to the development of novel therapies. Salmonella pathogenicity relies on its capacity to enter and replicate inside host cells. The process of invasion results in internalization of the bacterium within a membrane-bound compartment, the vacuole. The goal of this research proposal is to understand the molecular mechanisms of Salmonella intracellular survival. More specifically we are interested in gaining further insights into the function of late Salmonella effector proteins that are translocated from the bacteria into the host and which profoundly modify the host cell biology and insure a successful infection. We propose to integrate diverse disciplines ranging from microbiology to structural biology in order to create a holistic appreciation of the Salmonella pathogenicity. This comprehensive study will involve : 1-The development of biochemical and yeast two-hybrid screens for the identification of host protein partners of late Salmonella effectors; 2-Functional and biochemical analyses of the identified protein-protein complexes; 3-The resolution of the three-dimensional structures of the effectors (or functional domains) individually or in complex with the specific host protein partners. Three partners with complementary expertises in genetics, microbiology, cell biology and structural techniques will achieve these goals. Collectively, these considerations will participate to document the molecular mechanisms by which a selection of Salmonella late effectors interact with host proteins in order to manipulate host-cell processes and are expected to lead to the definition of novel therapeutic and preventive strategies.

Methods of estimation of additional mortality due to cancer aim at estimating and modelling the excess mortality to which a studied group of patient's cancer is subjected and at estimating their relative survival (i.e. the survival corrected for all the other causes of death). In this context, the objective of the modelling is to estimate the impact of prognostic factors on the excess mortality risk and to assess the cure rate in different subgroups of patients. These methods are used in order to produce statistics concerning survival of patients with cancer, estimated from population-based data collected by cancer registries. In France, population-based cancer registration is carried out on a Departmental level. These registries are grouped together in the FRANCIM network, for which one of the topic of its research program in cancer epidemiology focus on the estimation of the survival of cancer's patients in France. In order to achieve this, data of cancer registries have been pooled in a common database localised in the Service de Biostatistiques des Hospices Civils de Lyon. These statistics are analysed regularly and published by the different European countries and comparisons between countries are justified only if the methods used for that have taken into account bias relative of observational studies and if they are the result of a thought and a strategy adopted by all the partners. The development and the homogenisation of such methodology are totally justified in this context. The objective of this project is to provide a more complete set of analytic tools to estimate cancer survival and to assess the public health implications of these estimates. The research will produce a range of new clinical and public health insights from cancer survival data. We propose to explore and to address two complementary themes: - Improvements in cancer relative survival analysis methodology: extension of methods developed for censored data to the framework of relative survival analysis; - Creation of a 'standardized' approach for estimating cancer survival in France, or even in Europe, using Registry data. This project will allow for the creation of a network including 5 French partners, having complementarities and experience in the framework of the analysis of mortality due to cancer and the development of statistical methods. Two European partners and one Canadian, internationally known, will participate to this project as external members. The project is divided into 3 work packages: WP1: Comparison of the actual methods used to estimate relative survival WP2: Extension and development of statistical methods used to estimate relative survival. This necessitate (i) statistical researches to deal with the important remaining issues, and (ii) to develop statistical programs to perform relative survival analyses using approaches studied and developed in this project. WP3: Production of guidelines for cancer survival analysis with population-based data, answering to public health considerations The performances of the estimators of these new models will be assessed by simulations. Applications based on real data will be used to illustrate these new approaches and for the elaboration of guidelines. Data will be provided by some French cancer registries and by our European partners. Specific works are defined for each partner who will produce specific deliverables. If the project goes on well, it will allow to propose an adapted methodology in order to obtain correct estimates of the excess mortality due to cancer and to its determinants. This methodological approach is a preliminary condition for a rational management of such disease, on its medical and socio-economic aspects, that will be obtained from registries data.