Research

Our long-term goal is to understand how the nanostructure of lipid membranes controls biological function (e.g., cellular signaling). Once we understand the physical factors governing the lateral and dynamic heterogeneities in membranes, we will be able to relate how membrane structure controls molecular localization, thereby controlling cellular function. Our overall objective is to develop new analytical and physical tools to follow the dynamics of single molecules on biomimetic membranes with high spatial and temporal resolution.

Here are some of the projects going on in our group.

Localization of proteins and lipids and their specific interactions in biomembranes impact cellular function and are driven by local membrane structure. The fluid mosaic model suggests that the plasma membrane is primarily composed of a lipid bilayer in which integral membrane proteins are embedded. In the simplest interpretation of this model, lipids and proteins are randomly distributed in the membrane and are able to freely diffuse in the two-dimensional matrix. However, studies on lateral mobilities of membrane components suggested considerable revision of our understanding of the plasma membrane. It is now recognized that specialized regions, or domains, can occur on the cell surface, at least transiently. Recently cholesterol-rich "lipid rafts" have been shown to facilitate signaling in mast cells and T cells. The sizes, lifetimes and molecular compositions of these rafts remain a mystery, in large part due to the complexity of cells. A temporal heterogeneity that has been encountered frequently in biomembranes is anomalous diffusion. In contrast to simple Brownian motion which has a linear dependence of the mean squared displacement <x ²> on time t (that is, the familiar <x ²> = 4 Dt ), for anomalous diffusion, , where is the anomalous diffusion exponent that is less than unity. New generations of model systems that more closely mimic biomembranes are needed for a realistic and quantitative investigation of membrane dynamics and the physical and chemical mechanisms underpinning these variations. Only then can we begin to understand the in vivo biophysics of membranes and develop models for signaling that can be experimentally tested. This effort also requires technique development that can be used for molecular-level understanding of properties and functions on biomembranes and other soft surfaces.

When IgE-IgE receptors are crosslinked with multivalent ligand (or antigen), the first biochemical step is phosphorylation of the receptor by the Src family kinase Lyn, which is anchored to the inner leaflet with two acyl chains. This step is hypothesized to be mediated by lipid rafts in a cholesterol-dependent manner. Drawn by ED Sheets.

We are probing the chemical and physical mechanisms that lead to anomalous diffusion. Although this has been observed on live cells and in some model systems, we want to explore systematically the various mechanisms leading to this type of non-Brownian motion. We are working with the theorist Michael J. Saxton to tackle this problem, which has important functional consequences.

We also are interested in specialized cholesterol-enriched domains that cell biologists call "lipid rafts". To understand the molecular interactions leading to these nanostructures as well as their dynamics, we are developing new ways to pattern more complex model membranes and biomimetic membranes using microfabrication. In these studies, we will probe how rafts actually behave in complex membranes, which will help us understand how they control membrane function.

Lipid rafts facilitate the first biochemical step in the IgE receptor signaling pathway, which kicks off the allergic response in mast cells. We are developing anisotropy imaging to follow local lipid ordering in real-time as cells are stimulated. This should be a useful analytical tool to follow other types of signaling in cells.

Another area that we are working on is developing new analytical tools that will stimulate cells with well-defined spatial and temporal control. Again, we use microfabrication together with optical methods to control cell signaling. Using quantitative fluorescence microscopy, we will follow the molecular players as they participate in cell signaling. These tools should prove generally applicable to cell biology.

All of these projects use state-of-the-art optical microscopy to probe quantitatively the heterogeneous structure and dynamics of membranes. We've built a custom-designed multimodal imaging system that has differential interference contrast (DIC) and fluorescence capabilities as well as simultaneous CCD and PMT detection. Fluorescence excitation is done using either conventional epi-illumination with a mercury arc lamp or an argon ion laser. We also can selectively excite fluorophores that are <100 nm of a surface with evanescent field excitation, which results from total internal reflection of a laser that impinges on a surface at an angle greater than the critical angle.

An argon ion laser or HeNe lasers are used for (A) epi-illumination (pink path) and (B & C) evanescent illumination. A prism is optically coupled to the sample in the prism TIR path (B) . Not shown is conventional differential interference contrast (DIC). Optical trapping with a Ti:sapphire laser is show for a single trap (D) or for a dynamic holographic optical trap (E) configuration in a plane parallel to the laser-based epi-illumination. Detectors are an interline CCD camera for imaging and GaAs photomultipliers for dynamic fluorescence photobleaching recovery and correlation spectroscopy measurements. L: lens, M: mirror, DM: dichroic beamsplitter, FW: filter wheel. Drawn by Minjoung Kyoung.

In a traditional fluorescence photobleaching recovery (FPR) measurement, the average translational diffusion of fluorescently labeled molecules within the membrane is measured after irreversibly photobleaching the fluorophores with an intense laser pulse. The rate of fluorescence recovery (monitored with a much weaker laser beam) yields information about the lateral diffusion coefficient (D ) and the fraction of molecules that diffuse into the illumination volume on the time scale of the experiment. In contrast to FPR, single particle tracking (SPT) follows the movements of colloidal gold, fluorescent particles or molecules that are specifically bound to the membrane. These particles are tracked for a defined period of time using imaging to yield information about the local membrane environment on the submicron scale as well as an average D for the particle. FPR measures lateral diffusion of molecules at high concentration, which is the case for a range of biological systems. In special cases in which the diffusing molecules exist at very low concentrations, fluorescence correlation spectroscopy (FCS) is the technique of choice. In FCS, a low intensity laser beam is used to excite a fluorophore, at nanomolar concentrations, in a well-defined observation volume. Temporal fluctuations in the fluorescence are autocorrelated, and the shape of the resulting correlation function yields the characteristic diffusion time of the fluorophore as well as the average number of molecules in the observation volume. Furthermore, valuable information on any chemical kinetics that might be involved can be extracted. We also will use photon counting histogram (PCH) analysis to resolve a distribution of clusters or aggregates of fluorophores using the same data that is acquired in an FCS experiment. Finally, dynamic holographic optical trapping is used to create many traps in user-defined patterns from a single trapping laser. These individual traps can be uniquely and individually controlled, in real-time, to manipulate particles within any given pattern, and this technique is compatible with our other optical tools. Together, these analytical and physical techniques will allow us to quantitatively investigate the distributions and dynamics of molecules within membranes.

Filamentous action in fibroblasts were labeled specifically and imaged with wide-field excitation (red) or evanescent excitation (green). Note the focal adhesions that are selectively excited with the evanescent field. Figure courtesy of Minjoung Kyoung.