Intrinsically Disordered Proteins Sense Membrane Curvature

Abstract

melittin, appears to favor lipidation, where as for drugs, the depth of bilayer partitioning appears to be a crucial factor. In membranes with saturated lipids such as DPPC or DMPC, melittin undergoes little or no lipidation, whereas with unsaturated lipids such as POPC, modest lipidation is found. Cholesterol incorporation into all membrane types leads to striking increases in lipidation activity, alongside changes in the chemical selectivity of the transfer process. For some POPC/cholesterol mixtures, less than 50% of the original unmodified melittin is present after 24 hours. The level of activity does not correlate with the affinity of melittin for each membrane type. Rather the curvature modulus of the membrane is a better predictor of membrane activity. We hypothesise that the qualitative relationship between curvature modulus and lipidation activity is a consequence of changes in the penetration depth of melittin. Access of water to lipidation intermediates may also influence lipidation rates. 108-PlatIntrinsically Disordered Proteins Sense Membrane Curvature From endocytic vesicles to cytoskeletal protrusions, the ability of proteins to sense membrane curvature is essential for efficient initiation and assembly of curved membrane structures. To date, all known mechanisms rely on protein domains with specific structural features such as wedge-like amphipathic helices and crescent-shaped BAR domains. Many proteins that contain these structured domains also contain large intrinsically disordered regions. Nonetheless studies of curvature sensing have focused on structured domains in isolation, based on the assumption that curvature sensing requires structural order. In contrast, here we report that disordered domains are themselves potent sensors of membrane curvature. Comparison of Monte Carlo simulations with quantitative in vitro and live-cell measurements demonstrates that the polymer-like behavior of disordered domains found in endocytic proteins, including AP180, Epsin1, and Amphiphysin1, drives them to partition preferentially to convex membrane surfaces, which place fewer geometric constraints on their conformational entropy. For highly charged disordered proteins, electrostatic effects stiffen the peptide chain, reducing entropic curvature sensing. However, in these cases, increased repulsion between the disordered protein and the membrane surface give rise to an alternative, electrostatically derived mechanism of curvature sensing. Finally, full-length endocytic proteins, which contain both structured curvature sensors and disordered regions, are more than twice as curvature sensitive as their respective structured domains alone. These findings demonstrate curvature sensing mechanisms that are independent of protein structure and illustrate how structured and disordered domains can collaborate to synergistically enhance curvature sensitivity. MutSg plays a role in meiotic recombination facilitating crossover formation between homologous chromosomes. Failure to form crossovers leads to improper segre...

Molecular Mechanisms of Membrane Curvature Sensing by a Disordered Protein

“…27 Therefore, if curvature sensing arises from changes in conformational entropy, ΔG rel should scale linearly with N as AP180CTD is truncated, as demonstrated previously using Monte Carlo simulations. 14 This prediction is shown by the dashed line in Figure 1e. Interestingly, the measured reduction in ΔG rel with decreasing chain length is substantially weaker than the predicted reduction.…”

“…Here we demonstrate that an intrinsically disordered protein, the C-terminal domain of AP180, 14,[18][19][20] can sense membrane curvature using two distinct mechanisms. The first is entropic in nature.…”

“…Recently we reported that polymer-like disordered domains bind preferentially to small, highly curved vesicles 14 . This finding, which suggests that disordered proteins participate significantly in membrane curvature sensing, is highly significant owing to the abundance of disordered domains in membrane remodeling, [9][10][11] signaling, 15,16 and other membraneassociated functions.…”

“…To probe these mechanisms, here we examine a series of disordered amino acid chains of decreasing length, which consist of truncations of the C-terminal domain of the endocytic protein AP180 (AP180CTD). 14 Importantly, the physical 18,19 and biochemical 20 properties of AP180CTD have been well-studied, making this protein a good model for understanding the biophysical interactions between disordered proteins and membrane surfaces. For each truncation of AP180CTD, we quantified curvature sensitivity by measuring relative protein binding to substrate-tethered vesicles of increasing diameter.…”

The plasma membrane as a mechanochemical transducer

“…Some studies therefore claim that the sensing motif could be the ALPS motif only, with the driving force for its interaction with the PM being the density of defects rather than affinity [120]. How F-BARs or I-BARs that do not contain ALPS motifs would sense curvature is unclear, but crowding effects [121] or effects from the surrounding protein backbone [122] or the large disordered domains found in many of these proteins [123,124] could be involved. Most probably, the curvature sensing event is a cooperative process to which each of these domains contributes [125].…”