Preface to special issue on nanoscale membrane organisations.

The plasma membrane limits, protects and defines cell identity, separating two aqueous compartments: the cytoplasm from the extracellular space. However, cells need to transport nutrients, ions and waste products across their plasma membrane and also need to communicate with other cells. Cells intercommunicate usingmembrane receptors which transmit signals to the cytoplasm upon binding to their extracellular ligands. The extracellular ligands are soluble molecules or proteins embedded in the plasma membrane of other cells. In this way, the plasmamembrane has a double function: (i) as a barrier, constituted by a lipid bilayer, and (ii) as a communication platform, provided by protein receptors anchored to the lipid bilayer. The distribution and organisation of membrane receptors within the membrane is not irrelevant to their function, since receptors not only transmit outside– in information, a process coined as signal transduction, in the plane perpendicular to the plasma membrane, but can also influence each other in a horizontal manner, a process that can lead to cooperativity phenomena. A first model of plasma membrane organisation was proposed in 1972 by Singer and Nicolson who proposed that transmembrane proteins and lipids are homogeneously distributed and freely diffusing. This is known as the fluid mosaic model [1]. The validity of the fluid mosaic model was already challenged in the 1970s by Stier and Sackmannwhoproposed that the different chemico-physical properties of the lipids in the plasmamembranewouldmake them to self-organise in microdomains thus breaking the homogenous landscape of the fluid mosaic model [2]. The most prominent lipid microdomains are formed by cholesterol and sphingolipids, which self-associate forming a liquidordered phase (lipid raft), and separate from common phospholipids, which form a lipid-disordered phase. Consequent with this heterogeneous distribution of lipids Simons and Ikonen proposed thatmembrane proteins would also distribute non-homogenously and would have to partition either to the raft or to the non-raft phases [3]. The idea that membrane proteins are non-randomly distributed in membranes was supported by immuno-gold labelling electron microscopy studies on the distribution of receptors for immunoglobulins in mast cells [4]. However, the study of the distribution of antigen receptors in B and T lymphocytes [5,6] led to the idea that membrane proteins of the same type tend to concentrate in small areas of the membrane, forming oligomers of different size before the receptors are exposed to their extracellular ligands. Many receptors are now known to form nanometre-sized oligomers known as nanoclusters before ligand binding [5–9]. The distribution of membrane receptors in self-assembled nanoclusters is consequent with the general distribution of membrane proteins detected by electronmicroscopy and that led to the formulation of the protein island model in distinction, but not in contrast to, the lipid raft model [10]. Protein islands containing specific receptor nanoclusters