Our research focuses on the molecular mechanisms controlling glycemia in healthy and diabetic individuals. Diabetes is now considered as a worldwide epidemic by WHO, and is predicted to affect several hundred million people in the near future. Current therapies have failed to prevent or control hyperglycemia, as well as the deleterious cardiovascular consequences of the disease have increased. New paradigms are thus needed to develop novel therapeutic strategies. Over the last 15 years, we have been studying the intestine as a major regulator of the integrated cross-talk between the brain, liver, pancreas, muscles and blood vessels required for glycemic control. As a first example, we identified that during a meal the glucose transporter GLUT2 and the intestinal hormone glucagon-like peptide-1 (GLP-1) are involved in glucose detection by the entero-portal system. This was done using highly innovative experimental techniques in the awake free moving mouse. We then found that the enteric-vagal nervous system transmits this nutritional information towards the brain stem and hypothalamus, where leptin, neuropeptide Y and GLP-1 relay the enteric signal to control the endocrine pancreas (insulin-glucagon secretion), the liver (glycogen metabolism), the vascular system (vasodilation, arterial flow), and muscle metabolism. This "anticipatory metabolic reflex " is altered during diabetes and might thus represent a new pharmacological target. Subsequently, while investigating the molecular mechanisms responsible for regulating this new physiological pathway, we discovered that a fat-rich diabetogenic diet alters the intestinal microbiota and permeability. This leads to an increase in the concentration of plasma lipopolysaccharides (LPS), which causes metabolic endotoxemia responsible for the induction of low-grade inflammation that characterizes type 2 diabetes, insulin resistance, adipose tissue development and hepatic lipid storage. We then showed that bacteria can be translocated from the intestine towards tissues and the bloodstream. Bacterial DNA present in blood was found to be predictive of diabetes, 6-9 years before disease onset (patent), presenting new molecular targets in the microbiota-host relationship. This should enable the scientific community to discover new functional relationships between the genome and metagenome and thus to develop original preventive and therapeutic strategies for metabolic diseases. Four biotechnology companies have already been created on the basis of our findings.