The concept of transient nanometric domains known as lipid rafts has brought interest to reassess the validity of the Singer-Nicholson model of a fluid bilayer for cell membranes. However, this new view is still insufficient to explain the cellular control of surface lipid diversity or membrane deformability. During the past decade, the hypothesis that some lipids form large (submicrometric/mesoscale vs nanometric rafts) and stable (> min vs sec) membrane domains has emerged, largely based on indirect methods. Morphological evidence for stable submicrometric lipid domains, well-accepted for artificial and highly specialized biological membranes, was further reported for a variety of living cells from prokaryotes to yeast and mammalian cells. However, results remained questioned based on limitations of available fluorescent tools, use of poor lipid fixatives, and imaging artifacts due to non-resolved membrane projections. In this review, we will discuss recent evidence generated using powerful and innovative approaches such as lipid-specific toxin fragments that support the existence of submicrometric domains. We will integrate documented mechanisms involved in the formation and maintenance of these domains, and provide a perspective on their relevance on membrane deformability and regulation of membrane protein distribution.

Supplementary key words toxin • His-mCherry-NT-lysenin • lateral membrane heterogeneity • vital confocal imaging • membrane tension • cholesterol • temperatureLipids at the outer leafl et of the mammalian plasma membrane are mainly composed of: i ) SM, the most abundant sphingolipid (SL), based on a ceramide backbone and bearing a phosphocholine polar head; ii ) glycosphingolipids (GSLs), another group of SLs bearing various sugars instead of phosphocholine, from simple glucosylceramide (GlcCer) to complex GSLs such as GM1 [for a review, see ( 1 )]; iii ) phosphatidylcholine (PC), the major glycerophospholipid, sharing the same phosphocholine polar head as SM; and iv ) nonpolar cholesterol. Lipid bilayers are no longer considered as a homogenous solvent for membrane proteins ( 2 ), but are now represented with lateral heterogeneity at two different scales of time and space: i ) transient nanometric "lipid rafts", defi ned as small clusters enriched in SLs, sterol, and GPI-anchored proteins ( 3, 4 ); versus ii ) submicrometric/mesoscale domains ( 5-15 ). These larger and more stable domains are well-characterized on artifi cial vesicles ( 16, 17 ), but their relevance for living cells has been questioned ( 18,19 ).The occurrence in living cells of submicrometric/mesoscale domains was fi rst inferred from unexpected behavior in fl uorescence recovery after photobleaching (FRAP)

Extracellular vesicles (EVs) contribute to several pathophysiological processes and appear as emerging targets for disease diagnosis and therapy. However, successful translation from bench to bedside requires deeper understanding of EVs, in particular their diversity, composition, biogenesis and shedding mechanisms. In this review, we focus on plasma membrane-derived microvesicles (MVs), far less appreciated than exosomes. We integrate documented mechanisms involved in MV biogenesis and shedding, focusing on the red blood cell as a model. We then provide a perspective for the relevance of plasma membrane lipid composition and biophysical properties in microvesiculation on red blood cells but also platelets, immune and nervous cells as well as tumor cells. Although only a few data are available in this respect, most of them appear to converge to the idea that modulation of plasma membrane lipid content, transversal asymmetry and lateral heterogeneity in lipid domains may play a significant role in the vesiculation process. We suggest that lipid domains may represent platforms for inclusion/exclusion of membrane lipids and proteins into MVs and that MVs could originate from distinct domains during physiological processes and disease evolution.