Martin-Luther-Universität Halle-Wittenberg

Weiteres

Login für Redakteure

Research

Membrane remodeling by proteins

Eukaryotic cells have a very complex organization, featuring various, membrane-enclosed internal compartments. As a consequence, ‘cargo’ molecules need to be sorted and transported between these compartments. To facilitate transport, the formation of small vesicles from a donor lipid bilayer is mediated by coat proteins, which interact with the membrane. We are interested in how coat proteins accomplish the process of remodeling a lipid bilayer into a bud and finally into a separate vesicle. Important aspects of budding processes can be artificially reconstituted and studied in vitro using purified proteins and artificial membrane systems. Schematic view of the formation of a protein-coated vesicle from a lipid bilayer

Schematic view of the formation of a protein-coated vesicle from a lipid bilayer

Schematic view of the formation of a protein-coated vesicle from a lipid bilayer

Advantages of in vitro reconstitution

Protein chromatography

Protein chromatography

In vitro reconstitution represents a ‘bottom-up’ approach. A complicated biological process is taken apart and the various reactants can subsequently be examined by putting them together in any desired combination. To accomplish this, we express the proteins recombinantly, purify them and add them to artificial lipid bilayers. Among the advantages of using artificial membrane systems compared to biological cells is that we can control the composition of the lipids and choose among different types of spatial configurations. Moreover, artificial systems allow for a more flexible choice of biochemical, biophysical and microscopy techniques for characterizing the proteins and the lipid environment.

By studying the physicochemical properties of the proteins and their lipid environment and by exploring new avenues to reconstitution, we aim to make membrane proteins more accessible to high resolution structure determination by NMR and crystallography.

Artificial membrane systems

Liposomes can be prepared in various sizes, ranging from tens of nanometers to tens of micrometers. Reconstitution of integral membrane proteins into liposomes (i.e. the formation of proteoliposomes) as well as the peripheral binding of coat proteins to liposomes is analyzed using both biochemical and biophysical assays. (A) FCS (fluorescence correlation spectroscopy) analysis of membrane protein reconstitution. The protein carries a fluorescent label. Free protein in solution diffuses fast (red curve), whereas proteoliposomes diffuse much more slowly (purple curve). (B) Biochemical assay for membrane protein reconstitution.

(A) FCS (fluorescence correlation spectroscopy) analysis of membrane protein reconstitution. The protein carries a fluorescent label. Free protein in solution diffuses fast (red curve), whereas proteoliposomes diffuse much more slowly (purple curve). (B) Biochemical assay for membrane protein reconstitution.

(A) FCS (fluorescence correlation spectroscopy) analysis of membrane protein reconstitution. The protein carries a fluorescent label. Free protein in solution diffuses fast (red curve), whereas proteoliposomes diffuse much more slowly (purple curve). (B) Biochemical assay for membrane protein reconstitution.

We prepare very large liposomes, so-called Giant Unilamellar Vesicles (GUVs) by the electroformation method. GUVs are on the order of 10 µm in size, making them well-suited for studies by confocal fluorescence microscopy and fluorescence correlation spectroscopy (FCS, see below). (A) Electroformation setup for producing Giant Unilamellar Vesicle; (B) Dye-labeled Giant Unilamellar Vesicles in a test tube; (C) Confocal slice image of Giant Unilamellar Vesicles

(A) Electroformation setup for producing Giant Unilamellar Vesicle; (B) Dye-labeled Giant Unilamellar Vesicles in a test tube; (C) Confocal slice image of Giant Unilamellar Vesicles

(A) Electroformation setup for producing Giant Unilamellar Vesicle; (B) Dye-labeled Giant Unilamellar Vesicles in a test tube; (C) Confocal slice image of Giant Unilamellar Vesicles

Supported bilayers can be assembled on flat glass or mica surfaces for fluorescence imaging or AFM. We also prepare lipid bilayers on curved substrates to obtain membranes with a defined curvature. Langmuir monolayers are prepared at the buffer/air interface. In collaboration with the Blume lab, the interaction of proteins with the monolayers can be studied using infrared reflection absorption spectroscopy (IRRAS). (A) FRAP experiment on a supported planar bilayer containing fluorescently labeled lipid; (B) Supported spherical vesicles (SSVs)

(A) FRAP experiment on a supported planar bilayer containing fluorescently labeled lipid; (B) Supported spherical vesicles (SSVs)

(A) FRAP experiment on a supported planar bilayer containing fluorescently labeled lipid; (B) Supported spherical vesicles (SSVs)

(A) Schematic view of a lipid monolayer in a Langmuir balance. Protein is added to the subphase and changes in surface pressure are recorded. (B) Bound protein is analyzed by infrared reflection absorption spectroscopy (IRRAS).

(A) Schematic view of a lipid monolayer in a Langmuir balance. Protein is added to the subphase and changes in surface pressure are recorded. (B) Bound protein is analyzed by infrared reflection absorption spectroscopy (IRRAS).

(A) Schematic view of a lipid monolayer in a Langmuir balance. Protein is added to the subphase and changes in surface pressure are recorded. (B) Bound protein is analyzed by infrared reflection absorption spectroscopy (IRRAS).

Lipid phase behavior, protein-lipid interactions and Fluorescence Correlation Spectroscopy (FCS)
We are interested in the influence of the lipid environment on membrane proteins and vice versa. Dynamic lateral heterogeneities may have a role in membrane remodeling processes. This is a very attractive concept, but it has proven difficult to investigate experimentally. We use a combination of imaging techniques and single-molecule sensitive fluorescence spectroscopy to study membrane proteins and lipids in artificial systems. Using these methods we want to assess membrane protein reconstitution and learn about the intricate connection between lateral intramembrane sorting and vesicular budding. Budding in the absence of proteins. (A) Budding occurs along the separation line between the liquid-ordered phase (blue) and the liquid-disordered phase (pink). (B) The liquid-disordered phase and the liquid-ordered phase are characterized by distinct rates of probe diffusion, which can be determined by FCS. For details see Bacia et al., PNAS (2005). (C) SV40 virus-like particles (marked in red), which bind to the glycolipid GM1, are preferentially found on the liquid-ordered domain. The liquid-disordered phase is labeled in green. Virus sorting in vitro correlates with cellular uptake and infection in vivo. For details see Römer et al., Nature Cell Biology (2010).

Budding in the absence of proteins. (A) Budding occurs along the separation line between the liquid-ordered phase (blue) and the liquid-disordered phase (pink). (B) The liquid-disordered phase and the liquid-ordered phase are characterized by distinct rates of probe diffusion, which can be determined by FCS. For details see Bacia et al., PNAS (2005). (C) SV40 virus-like particles (marked in red), which bind to the glycolipid GM1, are preferentially found on the liquid-ordered domain. The liquid-disordered phase is labeled in green. Virus sorting in vitro correlates with cellular uptake and infection in vivo. For details see Römer et al., Nature Cell Biology (2010).

Budding in the absence of proteins. (A) Budding occurs along the separation line between the liquid-ordered phase (blue) and the liquid-disordered phase (pink). (B) The liquid-disordered phase and the liquid-ordered phase are characterized by distinct rates of probe diffusion, which can be determined by FCS. For details see Bacia et al., PNAS (2005). (C) SV40 virus-like particles (marked in red), which bind to the glycolipid GM1, are preferentially found on the liquid-ordered domain. The liquid-disordered phase is labeled in green. Virus sorting in vitro correlates with cellular uptake and infection in vivo. For details see Römer et al., Nature Cell Biology (2010).

Fluorescence correlation spectroscopy (FCS) is a very useful tool for examining mobility and interactions in a variety of systems including membranes. FCS is highly sensitive to small differences in the diffusion rates of proteins and lipids, which allows for instance to characterize differences in phase behavior of lipid bilayers. FCS is used to analyze the binding of diffusible ligands to membrane receptors, such as membrane proteins or glycolipids. Changes in the fluorescence brightness parameter reveal membrane protein oligomerization. Moreover, the use of dual-color fluorescence cross-correlation (dcFCCS) allows to assess protein-protein binding in cases, where binding does not lead to significant changes in diffusion rates. The dual-color cross-correlation technique can also be employed to detect dynamic co-localization of labeled cargo molecules in small, mobile carriers, such as transport vesicles. Owing to the use of fluorescent labels, FCS is highly specific and can be applied both to artificial, reconstituted systems and directly to living cells. Parameters accessible by FCS (for details see: Bacia et al., Nat. Methods 2006)

Parameters accessible by FCS (for details see: Bacia et al., Nat. Methods 2006)

Parameters accessible by FCS (for details see: Bacia et al., Nat. Methods 2006)

Schematic view of an FCS setup with dual color FCCS capability

Schematic view of an FCS setup with dual color FCCS capability

Schematic view of an FCS setup with dual color FCCS capability

FCS is typically performed on a setup that is similar to a confocal microscope. One or more laser lines are focused in the sample and the fluorescence is collected through the same objective. A pinhole serves to delimit the detection volume. The fluorescence emission(s) from the label(s) are selected by means of emission filter(s) and the fluorescence intensity as a function of time is recorded by avalanche photodiode detectors. Different methods of analysis are available to extract information from the fluorescence fluctuations, which occur as labeled molecule diffuse through the focus. Correlation analysis yields an autocorrelation curve, whose amplitude is inversely related to the concentration of the fluorescent particles. The decay time of the correlation curve reflects the diffusional mobility of the particles. In dual-color FCCS, the relative amplitude of the cross-correlation curve depends on the fraction of double-labeled (i.e., bound) particles.

Principle of FCS and dcFCCS measurements, for details see Bacia et al., Nat. Methods 2006

Principle of FCS and dcFCCS measurements, for details see Bacia et al., Nat. Methods 2006

Principle of FCS and dcFCCS measurements, for details see Bacia et al., Nat. Methods 2006

Direction and goals
We aim to develop methods for efficient reconstitution and enrichment of functional membrane protein for structural studies. We are especially interested in pharmaceutically relevant membrane receptors and enzymes.

Equipment
HALOmem is equipped with state-of-the art equipment for protein expression and purification (e.g. Äkta FPLC systems), protein analysis, crystallization (liquid handling robot) and physico-chemical analysis (differential scanning and isothermal calorimetry, FT-IR spectroscopy, dynamic light scattering, Langmuir balance). Fluorescence correlation spectroscopy (FCS) is used for single-molecule-sensitive diffusion and interaction analysis of membrane proteins and lipids in reconstituted membrane systems. Fluorescence auto- and cross-correlation measurements in combination with high sensitivity confocal fluorescence imaging can be performed on our Zeiss ConfoCor3/LSM710 setup. High speed confocal imaging is possible on our spinning-disk system. Cryo electron microscopy is carried out together with the university imaging facility.

Zum Seitenanfang