Molecular movements in biomembranes

Dr Tom Miller, Publisher of J. Phys. D: Appl. Phys., approached us two years ago with an idea of organizing a special issue on the application of single molecule techniques to processes in lipid membranes. Thinking about it, it was amazing to see how far the field has gone. Some 20 years ago, when first single molecule measurements on biomembranes [1] and biosystems in general [2] were reported, it would have been bold to predict the success story to come. But with improving microscopy technology and better fluorophores, experiments have left the highly specialized environments of physical chemists and are now routine also in biologically oriented labs. As a consequence, a wealth of novel findings has been reported and even reached the level of textbook knowledge.

Given the broad range of single molecule applications, we decided to focus here on molecular movements in biomembranes. We are happy that many renowned researchers accepted our invitation and contributed to this special issue. The choice of the topics was bottom-up: we had no guidelines, instead we wanted to see what our colleagues regarded as interesting enough to be included in this issue. It was thus exciting to observe convergences of the chosen topics to several main directions.

The heterogeneity of the cell plasma membrane on the sub-micrometer scale, including the presence of membrane microdomains [3], and the cytoskeleton-associated picket-fence protein meshwork [4], or, more generally, membrane rafts [5] created by the dynamic interplay between the local composition and phase of the membrane and membrane–protein interactions, has been a subject of intensive experimental and theoretical studies during the last decade. In spite of a considerable progress in understanding the underlying phenomena, there are still a lot of open questions in this field. These include the issues related to correct interpretation of experimental results on molecular movements in microheterogeneous membranes obtained with the well established experimental techniques, as well as development of new experimental approaches allowing one to address movements of membrane-associated molecules or colloidal particles at high spatial and temporal resolution. Four papers from this special issue address the effects of membrane microheterogeneity on mobility, distribution, and interactions of membrane-associated molecules, mainly focusing on correct interpretation of experimental results [6–9], whereas one more contribution [10] demonstrates the recent progress in an experimental technique for fast and accurate tracking of particles on lipid membranes. Šachl et al [6] and Burns et al [7] had a closer look onto spot-variation fluorescence correlation spectroscopy (svFCS), a technique introduced by the group of Didier Marguet ten years ago [11]. In this method, FCS autocorrelation curves recorded at different detection spot sizes are quantitatively compared, which allows for conclusions about the character of the underlying molecular motion. Experiments on cell membranes frequently yielded deviations from free Brownian motion, and conclusions were drawn on the presence of nanodomains to which fluorescent probe molecules preferentially partition [12]. Šachl et al analyzed the effects of movements of such nanodomains and found out that small domains can only be reliably detected if they are virtually immobile [6]. Burns et al found out that in case of critical fluctuations of the lipid membrane svFCS curves can give an impression of the presence of anomalous diffusion, even if the single molecule motion is nearly Brownian [7]. Related to this, Lagerholm et al [8] addressed the apparent contradiction between measurements of membrane protein and lipid mobility at ultrafast timescales performed with single particle tracking (SPT) versus stimulated emission depletion (STED) FCS with variable detection spot size. They concluded that the apparent contradiction can be successfully resolved when experimental artifacts in SPT data are properly accounted for. Arnold et al [9] studied the effects of deliberately arranged immobile obstacles in the plasma membrane on the mobility and the lateral distribution of probe molecules in order to establish a solid theoretical basis for quantitative interpretation of results obtained with a micropatterning technique recently put forward by the Schütz group [13]. The results of Arnold et al [9] provide an analysis framework for binding studies, yielding 2D dissociation constants from membrane probe mobility data. The paper by Spindler et al [10] expands the arsenal of experimental approaches for high-speed single molecule tracking at a high spatial localization accuracy by further developing the interferometric scattering (iSCAT) microscopy technique put forward by the Sandoghdar group [14]. The authors demonstrate the microheterogeneous character of supported lipid bilayers where they are able to track single lipids labeled with gold nanoparticles (GNP) as small as 5 nm (at the frame rate of ~20 kHz and localization precision of ~2 nm), or, alternatively, larger 20 nm diameter GNP at the record-beating frame rate approaching 1 MHz (at a localization precision better than 10 nm). The authors also demonstrate for the first time that iSCAT can be used for 3D tracking of nanoparticles attached to the membrane of a giant unilamellar vesicle. The paper additionally shows the capabilities of label-free iSCAT, which include monitoring the formation of a supported lipid bilayer, detecting and tracking single unlabeled proteins either performing Brownian motion or driven by an electric field along a flat substrate, as well as imaging of a cell membrane.

Interactions of colloidal (nano)particles and (bio)polymer molecules with lipid membranes govern diverse biologically relevant phenomena ranging from membrane–cytoskeleton interaction to cellular uptake of colloidal particles and macromolecules. Not surprisingly, a lot of efforts have been spent during the last decades to study this problem both experimentally and via computer simulations. Addressing these issues via computer simulations requires adequate description of the membrane dynamics, ranging from the local lipid demixing in multicomponent membranes to membrane elastic properties, preferably on experimentally relevant scales. In their topical review, Laradji et al [15] describe the successful application of an efficient coarse-grained implicit-solvent model [16] to membrane simulations. This approach allows one to correctly describe thermal, structural, and elastic properties of lipid membranes and efficiently address large-scale phenomena in lipid membranes ranging from phase separation in liposomes to membrane interactions with macromolecular structures and colloids, including cytoskeleton-induced blebbing, membrane wrapping of nanoparticles and the ensuing membrane-driven self-organization of colloids on the membrane, as well as endocytosis of nanoparticles by a tensionless membrane. The local membrane deformations and membrane-driven interactions play a crucial role in the interaction of negatively charged DNA macromolecules electrostatically bound to cationic lipid bilayers, which was experimentally addressed by Herold et al [17] using wide-field fluorescence video-microscopy using both freestanding and supported cationic lipid membranes in various phase states. They find that membrane-bound DNA macromolecules exhibit diverse conformational dynamics governed by the local properties of the lipid bilayer. In particular, in addition to the previously discovered membrane-driven irreversible collapse of DNA macromolecules on freestanding cationic lipid bilayers [18], Herold et al found different, fully reversible, compaction of DNA polymers on supported cationic lipid bilayers with fluid–gel phase coexistence [17]. In case where the colloidal particle can bind to the lipid membrane via dedicated lipid anchor groups, the particle-membrane interactions are mainly governed by the interplay of the particle shape and positioning of the anchor groups on the particle. Khmelinskaia et al [19] experimentally addressed such a problem by studying interactions of elongated parallelepiped-shaped DNA origami nanostructures with varying placement and number of cholesteryl membrane anchors. The authors found that at least two anchors positioned near the corners of the structure are required for strong and efficient binding of the DNA origami nanostructure to the membrane. A further increase in the number of anchors ensures stronger membrane binding of the nanostructures and slows down their Brownian motion on the membrane, as detemined using FCS measurements.

Finally, two papers further extended the scope to more physiological systems. Torreno-Pina et al [20] review recent results on SPT studies which reported the direct observation of molecular interactions in the cellular plasma membrane. Without doubt, time-constants of protein–protein interactions are key parameters to understand signaling at the plasma membrane mechanistically. Torreno-Pina et al, however, also point out the multitude of interactions with non-labeled proteins, which may be observable indirectly via the exotic diffusion behavior of fluorescently labeled proteins [20]. Struntz and Weiss [21] go even one step further and carry out a study of the diffusion behavior of phospholipase C in C. elegans embryos. They use single-plane illumination microscopy (SPIM) to confine the illumination volume and measure the mobility via SPIM-FCS, which yields diffusion maps of selected membrane regions of the embryo. Considerable variability of the diffusion coefficients indicates microheterogeneities of the organism.

Taken together, we believe that this special issue provides a timely overview over the current concepts and questions about the movements of membrane constituents. We wish you an exciting and entertaining read.

GJS acknowledges support by the Austrian Science Fund (FWF projects P26337-B21 and P 25730-B21), and by the Austrian Research Promotion Agency FFG (842379).