Rahul Chadda

Dynamic Organizing Principles of the Plasma Membrane that Regulate Signal Transduction: Commemorating the Fortieth Anniversary of Singer and Nicolson's Fluid-Mosaic Model

The recent rapid accumulation of knowledge on the dynamics and structure of the plasma membrane has prompted major modifications of the textbook fluid-mosaic model. However, because the new data have been obtained in a variety of research contexts using various biological paradigms, the impact of the critical conceptual modifications on biomedical research and development has been limited. In this review, we try to synthesize our current biological, chemical, and physical knowledge about the plasma membrane to provide new fundamental organizing principles of this structure that underlie every molecular mechanism that realizes its functions. Special attention is paid to signal transduction function and the dynamic aspect of the organizing principles. We propose that the cooperative action of the hierarchical three-tiered mesoscale (2-300 nm) domains--actin-membrane-skeleton induced compartments (40-300 nm), raft domains (2-20 nm), and dynamic protein complex domains (3-10 nm)--is critical for membrane function and distinguishes the plasma membrane from a classical Singer-Nicolson-type model.

Membrane mechanisms for signal transduction: The coupling of the meso-scale raft domains to membrane-skeleton-induced compartments and dynamic protein complexes

Virtually all biological membranes on earth share the basic structure of a two-dimensional liquid. Such universality and peculiarity are comparable to those of the double helical structure of DNA, strongly suggesting the possibility that the fundamental mechanisms for the various functions of the plasma membrane could essentially be understood by a set of simple organizing principles, developed during the course of evolution. As an initial effort toward the development of such understanding, in this review, we present the concept of the cooperative action of the hierarchical three-tiered meso-scale (2-300 nm) domains in the plasma membrane: (1) actin membrane-skeleton-induced compartments (40-300 nm), (2) raft domains (2-20 nm), and (3) dynamic protein complex domains (3-10nm). Special attention is paid to the concept of meso-scale domains, where both thermal fluctuations and weak cooperativity play critical roles, and the coupling of the raft domains to the membrane-skeleton-induced compartments as well as dynamic protein complexes. The three-tiered meso-domain architecture of the plasma membrane provides an excellent perspective for understanding the membrane mechanisms of signal transduction.

Raft-based interactions of gangliosides with a GPI-anchored receptor

Gangliosides, glycosphingolipids containing one or more sialic acid(s) in the glyco-chain, are involved in various important physiological and pathological processes in the plasma membrane. However, their exact functions are poorly understood, primarily because of the scarcity of suitable fluorescent ganglioside analogs. Here, we developed methods for systematically synthesizing analogs that behave like their native counterparts in regard to partitioning into raft-related membrane domains or preparations. Single-fluorescent-molecule imaging in the live-cell plasma membrane revealed the clear but transient colocalization and codiffusion of fluorescent ganglioside analogs with a fluorescently labeled glycosylphosphatidylinisotol (GPI)-anchored protein, human CD59, with lifetimes of 12 ms for CD59 monomers, 40 ms for CD59s transient homodimer rafts in quiescent cells, and 48 ms for engaged-CD59-cluster rafts, in cholesterol- and GPI-anchoring-dependent manners. The ganglioside molecules were always mobile in quiescent cells. These results show that gangliosides continually and dynamically exchange between raft domains and the bulk domain, indicating that raft domains are dynamic entities.

Archipelago architecture of the focal adhesion: Membrane molecules freely enter and exit from the focal adhesion zone

The focal adhesion (FA) is an integrin‐based structure built in/on the plasma membrane, mechanically linking the extracellular matrix with the termini of actin stress fibers, providing key scaffolds for the cells to migrate in tissues. The FA was considered as a micron‐scale, massive assembly of various proteins, although its formation and decomposition occur quickly in several to several 10 s of minutes. The mechanism of rapid FA regulation has been a major mystery in cell biology. Here, using fast single fluorescent‐molecule imaging, we found that transferrin receptor and Thy1, non‐FA membrane proteins, readily enter the FA zone, diffuse rapidly there, and exit into the bulk plasma membrane. Integrin β3 also readily enters the FA zone, and repeatedly undergoes temporary immobilization and diffusion in the FA zone, whereas approximately one‐third of integrin β3 is immobilized there. These results are consistent with the archipelago architecture of the FA, which consists of many integrin islands: the membrane molecules enter the inter‐island channels rather freely, and the integrins in the integrin islands can be rapidly exchanged with those in the bulk membrane. Such an archipelago architecture would allow rapid FA formation and disintegration, and might be applicable to other large protein domains in the plasma membrane.

Rac1 Recruitment to the Archipelago Structure of the Focal Adhesion through the Fluid Membrane as Revealed by Single-Molecule Analysis

The focal adhesion (FA) is an integrin‐based structure built in/on the plasma membrane (PM), linking the extracellular matrix to the actin stress‐fibers, working as cell migration scaffolds. Previously, we proposed the archipelago architecture of the FA, in which FA largely consists of fluid membrane, dotted with small islands accumulating FA proteins: membrane molecules enter the inter‐island channels in the FA zone rather freely, and the integrins in the FA‐protein islands rapidly exchanges with those in the bulk membrane. Here, we examined how Rac1, a small G‐protein regulating FA formation, and its activators αPIX and βPIX, are recruited to the FA zones. PIX molecules are recruited from the cytoplasm to the FA zones directly. In contrast, majorities of Rac1 molecules first arrive from the cytoplasm on the general inner PM surface, and then enter the FA zones via lateral diffusion on the PM, which is possible due to rapid Rac1 diffusion even within the FA zones, slowed only by a factor of two to four compared with that outside. The constitutively‐active Rac1 mutant exhibited temporary and all‐time immobilizations in the FA zone, suggesting that upon PIX‐induced Rac1 activation at the FA‐protein islands, Rac1 tends to be immobilized at the FA‐protein islands.

Genetically Encoding Single-Molecule Fluorophores into Ion Channels in Living Cells

We have developed a new approach to track protein dynamics in living cells on a single-molecule level with genetically encoded fluorescent non-canonical amino acids (ncAAs). The probes are based on amino acids with free primary amines that are chemically coupled to commercially available fluorophores, such as Cy3 and Cy5. Specific incorporation of fluorescent ncAAs was achieved by nonsense suppression in Xenopus laevis oocytes and confirmed by two-electrode voltage clamp of rescued ionic currents as well as by total internal reflection fluorescence microscopy detecting single fluorophores in the plasma membrane. Overall, the data show that these fluorescent ncAA were accepted as substrates for the X. laevis translation machinery. To date, we have encoded five fluorescent ncAAs into intracellular, extracellular and transmembrane regions of a CLC chloride channel or a voltage-gated sodium channel. Moreover, the variation in the linker length between the alpha carbon atom and the fluorescent moiety of the ncAA was explored by coupling to para-amino-phenylalanine, lysine and diaminopropionic acid and shows that linker length does not significantly affect the incorporation efficiency. This approach overcomes limitations of existing methods to study protein dynamics in living cells and is applicable to virtually any membrane protein. New applications include the detection of stoichiometry of macromolecular complexes by photo-bleaching of diffraction-limited puncta or the acquisition of relative distance measurements resulting from dynamic conformational changes within a membrane protein by Forster resonance energy transfer, both on a single-molecule level.

Measuring Large Membrane Protein Dimerization in Lipid Bilayers by Forster Resonance Energy Transfer

Our understanding of the thermodynamics forces in membrane protein assembly in a lipid bilayer is limited, because there are not many equilibrium models for studying protein association in lipid bilayers. We have developed a new model system based on the homodimeric CLC-ec1 Cl−/H+ antiporter (WT) that offers an opportunity to investigate this. A single tryptophan substitution (I422W) in WT yields a monomer/dimer mixture in detergent. We quantitatively labeled CLC-ec1 with Cy3 donor and/or Cy5 acceptor to measure Forster resonance energy transfer (FRET) changes upon dimerization. By single-molecule TIRF microscopy we measured the molecular FRET efficiency of as 0.7 ± 0.1 (n=3). We then measured the population FRET efficiency of I422W in large membranes while varying the protein/lipid fraction (χ) to determine a Kd of dimerization. Emission spectra obtained from the fluorometer is fit using an in-house software Fytt to decompose contributions from the sensitized emission of donor, acceptor, FRET and background and to quantify FRET efficiency (E). We determined E for the positive dimer control by co-labelling with Cy3/Cy5 across a range of χ. Measuring I422W-Cy3 + I422W-Cy5 in membranes shows low E at low χ that increases and saturates to all-dimer E at high χ, indicating a dimerization reaction in membranes. Furthermore, we observe reversibility of the reaction when going from a high χ to low χ by diluting all-dimer I422W membranes with empty lipid bilayers, and the appearance of FRET upon fusion of separately labelled I422W-Cy3 with I422W-Cy5 membranes, indicating subunit exchange. We compared 2:1 POPE/POPG vs. POPC lipid compositions, and while both show a similar reaction, the POPC data fits well to a dimerization isotherm with Kd(I422W) = 4 x 10−6 protein/lipid, suggesting that we are measuring equilibrium dimerization reaction in membranes.