Max Planck Fellow Group
Multifunctional Lipid Membranes on Surfaces

Prof. Dr. Claudia Steinem

For many years the chemical complexity as well as the dynamics and function of biological membranes have fascinated scientists. My group aims at generating robust and stable planar multifunctional membranes in a bottom-up approach that mimic the natural situation as closely as possible to address pressing biochemical questions and to design chip-based assays.

More than 40 years ago, biological membranes were described for the first time as a fluid mosaic based on thermodynamic principles of organization of membrane lipids and proteins and available evidence of asymmetry and lateral mobility within the membrane matrix. Over the intervening years, there is mounting evidence that the chemical and spatial complexity of biological membranes is key to understand their dynamics and functions on various length scales. It is the heterogeneity of cellular membranes that can lead to specialized functional membrane domains, enriched in certain lipids and proteins, and the interaction between lipids and proteins that limits the lateral diffusion and range of motion of membrane components thus altering their function. To be able to understand the complex interplay within a membrane on a molecular level, my research group pursues a bottom-up approach. By developing and applying model membrane systems, we aim to understand membrane-confined processes such as fusion and fission, transport processes mediated by ion channels and protein pumps as well as protein-lipid and protein-protein interactions occurring at the membrane interface. On the one hand, we use planar supported lipid bilayers (PSLBs) and vesicles, such as giant unilamellar vesicles (GUVs). On the other hand, we have developed functional lipid bilayers on highly ordered pore arrays. These so-called pore-spanning membranes (PSMs) suspend nanometer- to micrometer-sized pores in an aluminum or silicon substrate (see Figure). They separate two aqueous compartments and can hence be envisioned as an intermediate between supported and freestanding membranes.

PSLBs and GUVs: Protein-membrane and protein-protein interactions

Several proteins use specific lipid receptors to attach to the plasma membrane. We are interested in the molecular interaction between these membrane-confined receptors and proteins and how this interaction influences the overall membrane structure. In this context, we focus on phosphatidylinositol phosphate binding proteins such as ezrin and the focal adhesion kinase harboring a FERM-domain as well as collybistin binding via a PH-domain. Another major target are glycosphingolipids such as Gb3, which serve as specific receptors for bacterial toxins such as Shiga toxin.

Ezrin links the plasma membrane to the cytoskeleton in its active state. It gets activated by binding to PtdIns(4,5)P2 in the plasma membrane and phosphorylation of a threonine. We investigated the mode of ezrin binding and its activation by using PSLBs in combination with surface sensitive techniques such as reflectometric interference spectroscopy, and fluorescence and atomic force microscopy. We further analyzed the coupling of actin and actomyosin networks via ezrin to these membranes thus generating a minimal actin cortex. With such a system in hand, we are able to address the question how the ezrin-PtdIns(4,5)P2 interaction influences the architecture of actin and actomyosin networks and how this impacts the dynamics, and mechanical properties of the composite system.

Collybistin is an adaptor protein that is involved, together with the scaffold protein gephyrin, in the recruitment of GABAA receptors to the postsynaptic density of inhibitory synapses. Our in vitro studies showed that full-length collybistin gets activated via an interaction with the C-terminal part of neuroligin-2 being prerequisite for binding of the PH-domain of collybistin to different phosphatidylinositol phosphates. To completely assemble the structures found at the postsynaptic membranes of inhibitory synapses, the question needs to be addressed how gephyrin, collybistin and neuroligin-2 act in concert, a process that can be further illuminated by using PSLBs in combination with reflectometric interference spectroscopy and high-resolution microscopy techniques.

In collaboration with the group of Prof. Dr. Daniel B. Werz (TU Braunschweig), we investigate the influence of the fatty acid of the globoside Gb3 on the partitioning in liquid-ordered/liquid disordered coexisting membranes, which is key to understand the primary step of Shiga toxin internalization. Shiga toxin, produced by Shigella dysenteriae and Shiga toxin producing E. coli strains gets internalized into the cell after binding of the B-subunits to Gb3 embedded in the plasma membrane of the host. We have shown that the molecular structure of Gb3 greatly influences its membrane partition, as well as membrane (re)organisation after toxin binding.

PSMs: Membrane fusion and transport proteins

Membrane fusion processes, mediated by SNAREs, are a hallmark of eukaryotic life. We are especially interested in membrane fusion during neuronal exocytosis. A number of in vitro fusion assays with these proteins reconstituted in artificial membranes have been established in recent years. However, it has still been proven difficult to monitor intermediate states of the fusion process in a system that captures the essential features of the in vivo system. We develop and apply a reconstituted membrane system based on PSMs. PSMs are long-term stable and can be formed on open pore arrays as well as on cavities. These setups allow for a quantitative analysis of the different stages during fusion of a single vesicle, such as docking, intermediate states and full fusion by means of fluorescence microscopy in a time-resolved manner (see Figure).

As PSMs are produced from spreading GUVs, they should also enable us to reconstitute ion channels and protein pumps with an appropriately high protein density. This is particularly important for ion channels and protein pumps that do not transport sufficient ions to be detected by electrophysiological recordings on single freestanding membranes. However, classical methods to produce GUVs such as electroformation and gentle hydration do not allow for a reproducible reconstitution of transmembrane proteins in GUVs with high protein content. Thus, we search for and establish alternative methods to generate GUVs based on droplets produced in microfluidic devices. Recently, we have shown that these GUVs produced from dropletstabilized GUVs can also be spread on porous substrates to generate PSMs. With respect to proteins, we currently focus our attention on the F0F1 ATP-synthase that is capable of producing ATP by using a proton gradient and which can be reversed as well as connexons that can even connect two membranes by the formation of gap junctions.