PhD projects

ITP FP7 - SBMPs

Structural Biology of Membrane Proteins

PhD project:

Structure and function of the serotonin-gated ion channel 5HT3 receptor (part 1)




Principal Investigator:

Prof H. Vogel, Institut des Sciences et Ingénierie Chimiques, Ecole Polytechnique Fédérale de Lausanne (EPFL); horst.vogel|epfl.ch.


In collaboration with:

Prof. A. Engel, M.E. Müller Institute for Microscopy, Biozentrum, University of Basel, 4056 Basel, Switzerland; andreas.engel|unibas.ch

Prof. E. Pebay-Peyroula, Inst de Biol. Struct. UMR5075 CEA-CNRS-Univ. J.Fourier, 41, rue Jules Horowitz, F38027 Grenoble cedex 1; eva.pebay-peyroula|ibs.fr

Ionotropic receptors such as ligand-gated ion channels (LGIC) are central in the rapid communication within and between cells. They are activated within a few microseconds after the binding of agonists, leading to the opening of the transmembrane channel. Upon prolonged presence of the agonist, the receptor desensitizes within seconds, i.e. the LGIC enters in a long lasting state with the channel shut. Thus the receptor cannot be activated directly again by agonist binding. Because the ligand-binding site is located about 6 nm distant from the channel gate, it is evident that the receptor must undergo structural changes both in the ligand binding and in the transmembrane channel region (Unwin 2005).

The best studied LGIC is the muscle-type nAchR, which is the archaetype of the Cys-loop family comprising also the 5HT3R, the γ-aminobutyric acid type A and glycine receptors (Tompson 2006). More than 200 genes have been documented for this family. In this project we will concentrate on the serotonin-gated 5HT3R. Phylogenetic analysis has shown that the 5HT3R is one of the oldest members of the family, suggesting that it could serve as a general model for all Cys-loop receptors. As expression of 5HT3R subunit A alone leads to functional receptors with virtually native characteristics, the 5HT3R is an ideal model for LGICs: It functions as a homopentameric protein and carries the blueprint for the entire receptor family. 5HT3R plays a role in pain, mood, addiction, bradycardia and colonal motility (e.g. Kaumann 2006, and refs therein), and its antagonists are used to treat emesis due to anti-cancer chemotherapy and colonal disorders.

Presently the only published 3D structure of a LGIC is the 4 Å resolution EM-structure of the unliganded nAchR from Torpedo (Unwin 2005). Here we propose to investigate the structure of the 5HT3R: (i) Heterologous expression of the receptor, purification in detergent solubilized form, lipid reconstitution and functional characterization of the purified receptor will be the basis for starting structure determination. (ii) Electron  microscopy of detergent solubilized receptor and 2D crystallized receptor will deliver a first low resolution structure of the receptor. (iii) First steps towards 3D crystallizing the receptor protein will be the basis for elucidating the 3D structure at high resolution.


The group of H. Vogel has long-standing expertise in different biophysical techniques to investigate the structural dynamics of membrane proteins in particular also the 5HT3R (Schreiter 2003; Guignet 2004; Illegems 2004, 2005). Of central importance in the present project are the following recent findings: The EPFL group optimized heterologous expression of wild-type and mutant 5HT3R (Pick 2003) yielding on average a few million 5HT3R per cell. Fluorescence micrographs show patching of the over-expressed receptor in the plasma membrane of the cells with a density of about 10’000 receptors per square micrometer, which is comparable with receptor densities in 2D crystals (Pick 2003). This observation is very promising to start to produce 2D crystals of the 5HT3R for structural studies by electron microscopy. Furthermore, the EPFL group succeeded to purify functionally active, detergent-solubilized oligo-histidine containing 5HT3R in a single step using a Ni-NTA affinity column, yielding about 1 mg receptor per litre cell culture (Illegems 2005). This will be the basis to start trials to produce 3D crystals for X-ray crystal structure investigations.

 The group of A. Engel has a broad experience in determining the 3D structure of membrane proteins by electron microscopy (EM). The group has advanced the method of 2D crystallization and has more recently contributed software for efficient extraction of the pertinent information from micrographs or diffraction patterns of highly ordered 2D crystals. The expertise in the group has been applied to the study of aquaporins and different bacterial porins (Murata 2000; Remigy 2003; Kukulski 2005; Schenk 2005; Signorell 2007; Philippsen 2007).

The group of E. Pebay-Peyroula has outstanding expertise on determining the 3D structure of membrane proteins by X-ray crystallography (Nury 2006; Pebay-Peyroula).


This PhD project will focus on the following activities:


- Protein expression and purification (H. Vogel’s group, 9 months)

The protocols for the purification of the 5HT3R will be optimized for yield and purity: (i) Instead of using the standard Ni-NTA column we will investigate whether multidentate NTA colums (e.g. agrose comprising tris-NTA) deliver higher yield and purity of the the His-tagged 5HT3R. (ii)  An additional affinity purification step will be included (e.g. Strep-tag, biotin-tag, or affinity ligand) in order to obtain higher purity of the detergent solubilized purified receptor. (iii) Increasing protein expression by optimizing cell lines and growth conditions.


- Reconstitution into lipid bilayers and functional characterization of the 5HT3R (H. Vogel’s group, 9 months)

The homogeneity of the sample of the purified receptor will be investigated by fluorescence correlation spectroscopy which delivers information about size distribution of the micelles as well as the number of receptors per micelle; this is an important quality control for the subsequent 2D and 3D crystallization trials and will help to find the optimal detergent for purification. Protocols will be developed to reconstitute the purified 5HT3R into artificial lipid bilayers. The functionality of the receptor will be investigated by measuring binding of radioactive and fluorescent ligands, and by probing the receptor’s channel activity in planar lipid bilayers. Stability of the recepor in detergent and in lipid bilayers will be investigated by temperature- and urea/GdCl-induced denaturation/renaturation detecting circular dichroism, intrinsic as well as ligand fluorescence; during these experiments we expect to separately obtain information about the stability of the transmembrane and the extracellular ligand-binding part.


- Characterisation of detergent solubilized receptors by EM; preparation of 2D crystals and EM imaging (A. Engel’s group, 6 months)

Protein homogeneity will also be inspected by EM of negatively stained preparations. Such samples labelled with ligands attached to gold nanoparticles, which still bind with high affinity to the detergent-solubilized 5HT3R, will be studied by cryo-EM. We anticipate the nanogold label to facilitate single particle processing. Reconstitution experiments will start from conditions established by functional studies, but protein concentration will be increased to obtain a lipid-to-protein ratio (LPR) close to 1 to foster crystallization. Forcing the detergent concentration below the cmc will induce the reconstitution, and ultimately the 2D crystallization. Three methods will be explored: (i) dialysis if the 5HT3R is sufficiently stable in a relatively high CMC detergent (e.g. decyl maltoside), (ii) using cyclodextrin to remove the detergent stoichiometrically, and (iii) using BioBeads for detergent removal in case of difficult detergents that cannot be trapped by cyclodextrin but have too small a CMC for the dialysis method. Besides systematic variations of the LPR, we will explore pH, salts and cofactors (e.g., ligands). The result of 2D crystallization will be analysed by negative stain EM and, provided that excellent crystals are obtained, electron crystallography).


- Preparation of 3D crystals; X-ray crystallography (E. Pebay-Peyroula’s group, 12 months)

Stabilization conditions of the receptor solubilized in detergent prior crystallization will be deduced from the characterization done in Vogel’s group. In particular, these studies will guide the choice of the ligands. Crystallization will be approached by the classical vapor diffusion method and also the lipidic cubic phases. Diffraction tests on microcrystals at ESRF will guide the improvement of crystallization conditions. The structure will be solved by heavy atom derivatives. Depending on the success in crystallization, the structure of the receptor co-crystallized with different ligands could explored.

References:

  1. E. Guignet, R. Hovius, H. Vogel
    Reversible site-selective labeling of membrane proteins in live cells.  Nature Biotechnol. 22, 440-444, (2004).
  2. E. Ilegems, H. Pick, H. Vogel.
    Noninvasive imaging of 5-HT3 receptor trafficking in live cells: from biosynthesis to endocytosis. J. Biol. Chem.  279, 53346-53352, (2004).
  3. E. Ilegems, H. Pick, C. Deluz, S. Kellenberger, H. Vogel.
    Ligand binding transmits conformational changes across the membrane-spanning part to the intracellular side of the 5HT3 serotonin receptor. ChemBioChem 6, 2180-2185, (2005).
  4. AJ. Kaumann, FO. Levy
    5-Hydroxytryptamine receptors in the human cardiovascular system. Pharmacol. Therap. 111, 674-706, (2006).
  5. W. Kukulski, AD. Schenk, U. Johanson, T. Braun, BL. de Groot, et al.
    The 5A structure of heterologously expressed plant aquaporin SoPIP2; 1. J Mol Biol 350, 611, (2005).
  6. K. Murata , K. Mitsuoka, T. Hirai, T. Walz, P. Agre, et al.
    Structural determinants of water permeation through aquaporin-1. Nature 407, 599, (2000).
  7. H. Nury, C. Dahout-Gonzalez, V. Trézéguet, G. Lauquin, G. Brandolin, E. Pebay-Peyroula.
    Relations between structure and function of the mitochondrial ADP/ATP carrier. Annual Review of Biochemistry, 75, 713-741, (2006).
  8. E. Pebay-Peyroula, C. Dahout-Gonzalez, R. Kahn, V. Trézéguet, G.J.-M. Lauquin, G. Brandolin.
    Structure of mitochondrial ADP/AP carrier in complex with carboxyatractyloside. Nature 426, 39-44, (2003).
  9. A. Philippsen, AD. Schenk, GA. Signorell, V. Mariani, S. Berneche, A. Engel.
    Collaborative EM image processing with the IPLT image processing library and toolbox. J Struct Biol 157, 28, (2007).
  10. H. Pick, AK. Preuss, M. Mayer, T. Wohland, R. Hovius, H. Vogel.
    Monitoring expression and clustering of the ionotropic 5HT3 receptor in plasma membranes of live biological cells. Biochemistry  42, 877-884, (2003).
  11. HW. Remigy, D. Caujolle-Bert, K. Suda, A. Schenk, M. Chami, A. Engel.
    Membrane protein reconstitution and crystallization by controlled dilution. FEBS Lett 555, 160, (2003).
  12. AD. Schenk, PJ. Werten, S. Scheuring, BL. de Groot, SA. Muller, et al.
    The 4.5 A structure of human AQP2. J Mol Biol 350, 278, (2005).
  13. C. Schreiter, R. Hovius, M. Costioli, H. Pick, S. Kellenberger, L. Schild, H. Vogel.
    Characterization of the ligand-binding site of the serotonin 5-HT3 Receptor.  J. Biol. Chem. 278,, 22709-22716, (2003).
  14. GA. Signorell, TC. Kaufmann, W. Kukulski, A. Engel, HW. Remigy.
    Controlled 2D crystallization of membrane proteins using methyl-beta-cyclodextrin. J Struct Biol 157, 321, (2007).
  15. AJ. Thompson, SCR. Lummis.
    5-HT3 receptors. Curr. Pharmaceut. Design. 12, 3615-3630, (2006).
  16. N. Unwin.
    Refined structure of the nicotinic acetylcholine receptor at 4 Angstrom resolution. J Mol Biol 346, 967-989, (2005).

ITP FP7 - SBMPs

Structural Biology of Membrane Proteins

PhD project:

Structure and function of the serotonin-gated ion channel 5HT3 receptor (part 2)




Principal Investigator:

Prof H. Vogel, Institut des Sciences et Ingénierie Chimiques, Ecole Polytechnique Fédérale de Lausanne (EPFL); horst.vogel|epfl.ch.


In collaboration with:

Prof. A. Engel, M.E. Müller Institute for Microscopy, Biozentrum, University of Basel, 4056 Basel, Switzerland; andreas.engel|unibas.ch.

Prof. D. Muller Center of Biotechnology, University of Technology Tatzberg 47-51, D-01307 Dresden, Germany; mueller|biotec.tu-dresden.de.

Ligand gated ion channels (LGIC) are activated within a few microseconds after the binding of agonists, opening a transmembrane channel, which allows the passage of ions along their electrochemical gradient. Upon prolonged presence of the agonist, the receptor desensitizes within seconds, i.e. the LGIC enters in a long lasting closed channel state. Thus the receptor cannot be activated directly again by agonist binding. In the presently best studied LGIC, the nicotinic acetylcholine receptor (nAchR), the ligand-binding site is located about 6 nm distant from the channel gate (Unwin 2005), which indicates that the receptor must undergo structural changes both in the ligand binding and in the transmembrane channel region.

The nAchR belongs to the so-called Cys-loop family comprising also the serotonin 5HT3R, the GABA and the glycine receptors (Tompson 2006). Here we concentrate on the 5HT3R because it is a general model of the Cys-loop family of LGIC and is ideally suited for structural studies: It functions as a homopentameric protein and carries the blueprint for the entire receptor family.

Here we propose to investigate the structure of the 5HT3R both in the unliganded form and in the presence of bound agonists and antagonists in native cellular membranes using (single molecule) fluorescence microscopy & electrophysiology, electron microscopy and atomic force microscopy (AFM) and single-molecule force spectroscopy (SMFS).


The group of H. Vogel has long-standing experience to investigate the structural dynamics of the 5HT3R by different biophysical techniques (Schreiter 2003; Guignet 2004; Illegems 2004, 2005). Of importance in the present project are the following recent findings:

(i) The EPFL group optimized heterologous expression of wild-type and mutant 5HT3R (Pick 2003) yielding on average a few million 5HT3R per cell. Fluorescence micrographs show patching of the over-expressed receptor in the plasma membrane of the cells with a density of about 10’000 receptors per square micrometer, which is comparable with receptor densities in 2D crystals (Pick 2003). This observation is very promising to start to image the receptor by AFM to produce from native membrane sheets by detergent extraction of lipids 2D crystals of the 5HT3R for structural studies by electron microscopy.

(ii) Recent studies in the Vogel group have shown that agonist binding on the extracellular receptor side activates the channel and transmits structural changes to the intracellular portion of the 5HT3R, which can be monitored by optical probes inserted in the intracellular receptor part (Illegems 2005). This opens the possibility to study in detail structural changes of the receptor during channel activation.

(iii) In addition, the Vogel group has developed techniques to transfer single plasma membrane sheets from live cells to solid supports for investigating receptors by combined single molecule microscopy & planar patch clamp techniques (Perez 2006a, b; Danelon 2006; Schmidt 2000).


The group of A. Engel has a broad experience in determining the 3D structure of membrane proteins by electron microscopy (EM). The group has solved several large membrane bound complexes to 12-25 Å resolution by single particle analysis of projections from vitrified samples. The group has advanced the method of 2D crystallization and has more recently contributed software for efficient extraction of the pertinent information from micrographs or diffraction patterns of highly ordered 2D crystals. The expertise in the group has been applied to the study of aquaporins and different bacterial porins (Murata 2000; Remigy 2003; Kukulski 2005; Schenk 2005; Signorell 2007; Philippsen 2007).


The group of D. Müller has outstanding expertise on high-resolution imaging membrane proteins by AFM and recently developed efficient approaches to characterise the unfolding and refolding pathways and energy landscape of membrane proteins by SMFS (Müller 2006; Kedrov 2007). These techniques have been applied so far on various membrane proteins, and deserve to be applied on 5HT3R in native plasma membrane sheets.


This PhD project will focus on the following activities:


- Preparation and functional characterization of native membranes comprising densely packed 5HT3R (H. Vogel’s group, 18 months)

Based on the finding of Pick et al (2003), the 5HT3R will be expressed up to 10 million copies per cell (HEK, CHO). Under these conditions the receptor forms large patches with high receptor densities (10’000 rec/µm2). Two different membrane preparations will be used for combined fluorescence & electrophysiological experiments and for AFM and EM imaging: (i) Membrane patches from the cellular plasma membranes will be isolated by density centrifugation. Here protocols will be developed for obtaining pure membrane fractions with highest receptor densities. (ii) Alternatively, membrane fragments will be transferred directly from the live cells to solid supports using a method developed by the Vogel group recently (Perez 2006 a, b; Danelon 2006). The two different preparations will be first characterized for their biological function (ligand binding, channel activation) and will be then either used at the EPFL for fluorescence/electrical experiments or transferred to the partner labs.

At the EPFL the planar membrane sheets will be transferred to different solid supports, preferentially to silicon chips which carry one or multiple (sub)micrometer-sized holes and are covered on one surface by a 100 nm film of Al-oxide (Danelon 2006). In this configuration the chip functions as a mono-mode wave guide: Light waves addressed from one side towards the chip will create a high-intensity evanescent wave, concentrated in the holes, suited to probe fluorophores in the planer membrane assembled across these holes; fluorescent probes might arise from fluorescent ligands bound to the 5HT3R (Pick 2003, Illegems 2004) or from probes directly integrated in the receptor in form of GFP (Illegems 2005), as a labeled ACP-tag as pioneered by the EPFL group for imaging membrane proteins (Meyer 2006; Jacquier 2006), or 5HT3R mutant proteins selectively labeled on polyhistidine sequences (Guignet 2004). Membranes, assembled on one side of the chip will form an electrical isolating layer. With two electrodes placed in microliter volumes on both sides of the chip, ionotropic receptors in the planar membranes can be characterized by electrical ion current measurements, if necessary with single channel resolution (Schmidt 2000; Danelon 2006). We will use this chip platform to measure simultaneously fluorescence signals and electrical ion channel currents. This approach will deliver important novel information about the structural changes occurring during opening and closing of the ion channels. In a second series of experiments, we will dilute the 5HT3R density in the planar membranes by fusing native membrane fragments with artificial lipid bilayers in order to do single molecule fluorescence/electrical channel measurements to obtain details of the structural fluctuations within a single LGIC. The studies will be complemented by EM and AFM structural work (below).


- EM imaging plasma membrane sheets of high 5HT3R density and 5HT3R-2D crystals induced by phospholipase treatment of plasma membranes (A. Engel’s group, 9 months)

Highly enriched plasma membranes bearing the recombinant 5HT3R will be analyzed by negative stain as well as cryo-EM to determine the packing density of 5HT3R. Such membranes will be treated by phospholipase A2 in the presence of cyclodextrin to extract interspersed lipids, thereby increasing the receptor’s packing density. By this approach crystallization will be induced. Variation of pH and salts will be explored to improve the crystal quality. The process will be monitored by negative stain EM. After optimization of the conditions, samples will be studied by electron crystallography.


- AFM: Imaging and interactions of 5HT3R in plasma membrane sheets (D. Müller’s group, 9 months)

Using AFM the PhD student will image the 5HT3R in their native plasma membrane sheets at high-resolution. AFM will enable to observe the oligomeric arrangement of the 5HT3R subunits and the 5HT3R assembly in the native membrane. Functional AFM imaging will provide insights into possible conformational changes of 5HT3R. Additionally SMFS will provide a quantitative view of the energetic landscape for the folding and unfolding of the 5HT3R.  In a first step we will detect and map molecular interactions within 5HT3R. Dynamic SMFS will then map details of the energy landscape and observe the energy minima, energy barriers and kinetic properties of the 5HT3R structure.

  1. C. Danelon, JB. Perez, H. Vogel.
    Cell membranes suspended across nano-aperture arrays. Langmuir 22, 22-25, (2006).
  2. E. Guignet, R. Hovius, H. Vogel.
    Reversible site-selective labeling of membrane proteins in live cells. Nature Biotechnol 22, 440-444, (2004).
  3. E. Ilegems, H. Pick, H. Vogel.
    Noninvasive imaging of 5-HT3 receptor trafficking in live cells: from biosynthesis to endocytosis. J Biol Chem 279, 53346-53352, (2004).
  4. E. Ilegems, H. Pick, C. Deluz, S. Kellenberger, H. Vogel.
    Ligand binding transmits conformational changes across the membrane-spanning part to the intracellular side of the 5HT3 serotonin receptor. ChemBioChem 6, 2180-2185, (2005).
  5. V. Jacquier, M Prummer, JM. Segura, H. Pick, H. Vogel.
    Visualizing odorant receptor trafficking in living cells down to the single-molecule level. Proc Natl Acad Sci USA 103, 14325-14330, (2006).
  6. A. Kedrov, et al.,
    Deciphering molecular interactions of native membrane proteins by single-molecule force spectroscopy. Annu Rev Biophys Biomol Struct 36, 233-260, (2007).
  7. W. Kukulski, et al.,
    The 5A structure of heterologously expressed plant aquaporin SoPIP2; 1. J Mol Biol 350, 611, (2005).
  8. BH. Meyer, JM. Segura, KL. Martinez, R. Hovius, N. George, K. Johnsson, H. Vogel.
    FRET imaging reveals that functional neurokinin-1 receptors are monomeric and reside in membrane microdomains of live cells. Proc Natl Acad USA 103, 2138-2143, (2006).
  9. DJ. Muller, et al.,
    Single-molecule studies of membrane proteins. Curr Opin Struct Biol 16, 489-495, (2006).
  10. K. Murata, et al.,
    Structural determinants of water permeation through aquaporin-1. Nature 407, 599, (2000).
  11. JB. Perez, KL. Martinez, JM. Segura, H. Vogel.
    Supported cell membrane sheets for functional fluorescence imaging of membrane proteins. Adv Funct Mat 16, 306-312, (2006a).
  12. JB. Perez, JM. Segura, KL. Martinez, D. Abankwa, H. Vogel.
    Monitoring the diffusion of single heterotrimeric G proteins in supported cell-membrane sheets reveals their partitioning into microdomains. J Mol Biol 363, 918-930, (2006b).
  13. A. Philippsen, AD. Schenk, GA. Signorell, V. Mariani, S. Berneche, A. Engel.
    Collaborative EM image processing with the IPLT image processing library and toolbox. J Struct Biol 157, 28, (2007).
  14. H. Pick, AK. Preuss, M. Mayer, T. Wohland, R. Hovius, H. Vogel.
    Monitoring expression and clustering of the ionotropic 5HT3 receptor in plasma membranes of live biological cells. Biochemistry 42, 877-884, (2003).
  15. HW. Remigy, D. Caujolle-Bert, K. Suda, A. Schenk, M. Chami, A. Engel.
    Membrane protein reconstitution and crystallization by controlled dilution. FEBS Lett 555, 160, (2003).
  16. AD. Schenk, et al.,
    The 4.5 A structure of human AQP2. J Mol Biol 350, 278, (2005).
  17. C. Schmidt, M. Mayer, H. Vogel.
    A chip-based biosensor for functional analysis of single ion channels. Angew Chemie Int Ed 39, 3137-3140, (2000).
  18. C. Schreiter, R. Hovius, M. Costioli, H. Pick, S Kellenberger, L. Schild, H. Vogel.
    Characterization of the ligand-binding site of the serotonin 5-HT3 Receptor. J Biol Chem 278, 22709-22716, (2003).
  19. GA. Signorell, TC. Kaufmann, W. Kukulski, A. Engel, HW. Remigy.
    Controlled 2D crystallization of membrane proteins using methyl-beta-cyclodextrin. J Struct Biol 157, 321, (2007).
  20. AJ Thompson, SCR Lummis.
    5-HT3 receptors. Curr. Pharmaceut. Design. 12, 3615-3630, (2006).
  21. N. Unwin.
    Refined structure of the nicotinic acetylcholine receptor at 4 Angstrom resolution. J Mol Biol 346, 967-989, (2005).