Several studies of the intercellular synapses between immune cells suggest that mechanical bending of the membrane can drive protein sorting and influence signal transduction events. Such bending effects can manifest at the molecular level, in which intermembrane protein complexes of differing sizes become mutually repulsive as a result of the mechanical deformations of the membrane they induce (e.g. Qi, Groves, and Chakraborty, Proc. Natl. Acad. Sci. USA 2001, 98, 6548; Weikl, Groves, Lipowsky, Europhys. Lett. 2002, 59, 916). Membrane bending can also mediate force transmission between membrane domains over distances of microns. Intrigued by the prospects of fluid cell membranes behaving as mechanical force transducers in cells, my group has developed an array of model membrane systems and complimentary imaging techniques to explore these phenomena.

The ability to image membrane topographical features is critical to understanding reaction mechanisms and organizational principles in cell membranes. We have introduced two strategies for imaging nanometer – scale topographical features in reconstituted membrane junctions. The first is based on intermembrane Förster resonance energy transfer (FRET) (J. Am Chem. Soc. 2001; Proc. Natl. Acad. Sci. USA 2002). Our quantitative studies of glycolipids and proteins with well-known structures have demonstrated that intermembrane FRET can resolve membrane spacing with Angstrom precesion. The second topographical imaging strategy is based on optical standing wave interferometry. This enables real-time mapping of the membrane surface with nanometer precision and provides resolution extending hundreds of nanometers from the surface. Using this system, we directly resolve thermal fluctuations and intrinsic topography of the membrane surface (J. Phys. Chem. B 2004; Cell Biochem. Biophys. 2004; Biophys. J. 2004). These observations provide insights into the characteristics of the membrane environment and offer new methods for studies of protein interactions within the membrane. Significantly, we have observed in cell free model systems spontaneous pattern formation driven by some the same mechanical forces suspected to be at play in living T cells (Proc. Natl. Acad. Sci. USA 2004; Phys. Rev. Lett. 2005; J. Phys Chem. B 2006) (Figure 2).

Membrane topography imaging by fluorescence interference has also proven useful for dynamical observations of membrane bending fluctuations. Using this technique to perform spatiotemporal thermal fluctuation spectroscopy, we observe strong hydrodynamic damping of the fluctuation time-scale (~10,000-fold) when the membranes are near an interface (Figure 3). Quantitative comparison between these measurements and a theoretical description of confined hydrodynamic effects has provided experimental confirmation of the model (Phys. Rev. Lett. 2006). Such hydrodynamic damping effects are likely to exist in junctions between living cells, where they may dominate the binding kinetics of membrane receptor proteins.

The prospect of membrane bending as a mechanism of long range force transduction could be very important in living cells (see, for example, the appended review manuscript: Parthsarathy and Groves). Based on the observation that phase separated membranes in the form of giant unilamellar vesicles (GUVs) are not spherical, we sought to develop more refined systems that would enable quantification of curvature effects in phase separated bilayer membranes. At one level, allowing a GUV to adhere to a substrate provided enough stabilization for highly ordered superstructures, such as stripes and hexagonal domain arrays, to form. The long range forces leading to the superstructure are curvature mediated (J. Am. Chem. Soc. 2005, 127, 36). Further clarification of curvature mediated forces between domains was provided by our introduction of curvature modulated substrates as a means of imposing a precisely defined curvature gradient to the membrane. Using this system, the bending rigidity difference between two coexisting liquid phases could be measured (Langmuir 2006). This allows rough estimation of the strength of curvature coupling in cell membranes and suggests that membrane bending (e.g. by forces applied from the cytoskeleton) would be an effective way of directly moving domains of clustered proteins around the membrane surface.

We have also used measurements of molecular diffusion in membranes, by fluorescence recovery after photobleaching (FRAP) and fluorescence cross correlation spectroscopy (FCS), to probe membrane structures. In this branch of work, emphasis has been directed on impact of protein binding to membrane surfaces on the lipid bilayer structure. Anomalously large changes in lipid diffusion have been detected near phase transitions. This allows diffusion measurements to be utilized as a readout of surface binding for analytical purposes (J. Am. Chem. Soc. 2005, 127, 2826). We have also combined high precision mobility measurements, by FCS, with attenuated total reflection Fourier transform infrared spectroscopy to simultaneously monitor membrane phase and mobility (Forstner, Lee, Parikh, and Groves manuscript appended).