Characterization of vinculin´s lipid
anchor region
Der Naturwissenschaftlichen Fakultät
der Friedrich-Alexander-Universität Erlangen-Nürnberg
zur
Erlangung des Doktorgrades
vorgelegt von
Gerold Diez
aus Bad Kissingen
Als Dissertation genehmigt von der Naturwissenschaftlichen
Fakultät der Universität Erlangen-Nürnberg
Tag der müdlichen Prüfung: 18.02.2009
Vorsitzender der
Promotionskommission: Prof. Dr. Eberhard Bänsch
Erstberichterstatter: Prof. Dr. Wolfgang H. Goldmann
Zweitberichterstatter: Prof. Dr. Hojatollah Vali
Partial results from this thesis were published in Biophysical and Biochemical Research
Communications. The following publication and conference contributions contain scientific
results of the thesis presented here.
Publications Mierke CT, Kollmannsberger P, Paranhos Zitterbart D, Raupach C, Diez G, Koch TM, Fabry B, Wolfgang H. Goldmann. Vinculin enhances cell invasion by increasing contractile force generation. Submitted to MCB 2009 Diez G, Kollmannsberger P, Koch TM, Mierke CT, Vali H, Fabry B, Goldmann WH. Anchorage of vinculin to lipid membranes influences cell mechanical behaviour. Submitted to Biophysical Journal (under revision) 2009 Möhl C, Kirchgeßner N, Schäfer C, Küpper K, Diez G, Goldmann WH, Merkel R, Hoffmann B; Modulation of vinculin exchange dynamics regulates adhesion site maturation and adhesion strength. Submitted to Cell Motility and Cytoskeleton (under revision) 2009 Klemm AH, Diez G, Alonso JL, Goldmann WH. Compareing the mechanical influence of vinculin, focal adhesion kinase and p53 in mouse embryonic fibroblasts. BBRC, Vol. 379, Page 799, Feb 2009 Diez G, List F, Smith J, Ziegler WH, Goldmann WH. Direct evidence of vinculin tail - lipid membrane interaction in beta-sheet conformation. BBRC Vol. 373,Page 69, Jun 2008 Schewkunow V, Sharma PS, Diez G, Klemm AH, Sharma PC, Goldmann WH. Thermodynamic evidence of non-muscle myosin II-lipid-membrane interaction. BBRC Vol. 336, Page 500, Feb 2008 Smith J, Diez G, Klemm AH, Schewkunow V, Goldmann WH CapZ-lipid membrane interactions: a computer analysis. Theor. Biol. Med. Model. Vol. 3 Page 30, Aug 2006 Scott DL, Diez G, Goldmann WH. Protein-Lipid Interactions: Correlation of a predictive algorithm for lipid-binding sites with three-dimensional structural data. Theo. Biol. Med. Model. Vol. 3, Page 17, Mar 2006 Conference contributions Diez G, Kollmannsberger P, Koch TM, Zitterbart DP, Mierke CT, Krukiewicz AA, Vali H, Fabry B, Goldmann WH. Anchoring of vinculin to the lipid membrane influences its tyrosine phosphorylation on position 1065. Meeting of the ASCB, San Fransisco, December 2008 Diez G, List F, Smith J, Himmel M, Ziegler WH, Goldmann WH. Characterization of vinculins membrane binding anchor-an in vitro study. 31st Annual Meeting of the DGZ (Marburg), EJCB Vol.87S1, Suppl.58, March 2008
Diez G, Kollmansberger P, Paranhos-Zitterbart D, Mierke CT, Smith J, Goldmann WH. Head and tail interactions of vinculin influence cell mechanical behaviour. 47.th Annual Meeting of the ASCB, Washington, December 2007
Diez G, Kollmansberger P, Goldmann WH. Anchoring of vinculin to lipid membranes influences its binding strength in living cells. 2nd European Meeting on Cell Mechanics, Barcelona September 2007
Diez G., Kollmansberger P., Vali H., Goldmann W.H. Anchoring of Vinculin to the membrane influences its binding strength in living cells. Annual CBB Meeting, McGill University Montreal (Canada), May 2007 Diez G., Smith J., Stiebritz M., Goldmann W.H. Molecular dynamics and secondary structure behaviour of the C-terminus of vinculin that includes a membrane binding anchor. German Physical Society Spring Meeting, Regensburg, Mar 2007 Diez G., Smith J., Goldmann W.H. Secondary structure computer analysis of vinculin’s C-terminus. 51st Biophysical Society Symposium, Mar 2007 Diez G., Scott D.L., Goldmann W.H. Protein-Lipid interactions: Correlation of a predictive algorithm for lipid-binding sites with 3D-structural data. Journal of Biomechanics, Vol. 39 (Sup.) P. 578, Aug 2006
Table of contents
1 Introduction ...................................................................................................................... 1
1.1 The extracellular matrix determines the different tissues .......................................... 1 1.2 Mechanical perturbations regulate cell survival and tissue formation....................... 3 1.3 The focal adhesion complex....................................................................................... 6 1.4 Focal adhesion formation ........................................................................................... 8 1.5 Vinculin a key player of the focal contact................................................................ 10
1.5.1 Lipid binding of vinculin.................................................................................. 12 1.6 Aims ......................................................................................................................... 14
2 Materials and Methods .................................................................................................. 15 2.1 Materials................................................................................................................... 15
2.1.1 Chemicals and enzymes ................................................................................... 15 2.1.2 Cell culture medium and plastic ware .............................................................. 15 2.1.3 Oligonucleotides............................................................................................... 15 2.1.4 Vectors ............................................................................................................. 16 2.1.5 Bacterial cultures and cell lines........................................................................ 16 2.1.6 Antibodies ........................................................................................................ 17 2.2.1 Methods in molecular biology.......................................................................... 17
2.2.1.1 Quantification of Desoxyribonuclein acid (DNA) ....................................... 17 2.2.1.2 DNA electrophoresis .................................................................................... 17 2.2.1.3 Restriction endonuclease digestion of DNA ................................................ 18 2.2.1.4 Ligation ........................................................................................................ 18 2.2.1.5 Preparation of chemical competent DH5α ................................................... 18 2.2.1.6 Transformation of competent cells............................................................... 19 2.2.1.7 Polymerase chain reaction (PCR) ................................................................ 19 2.2.1.8 Site directed mutagenesis ............................................................................. 19 2.2.1.9 Purification of plasmid DNA from bacteria................................................. 20 2.2.1.10 DNA sequencing ...................................................................................... 20 2.2.1.11 DNA extraction from agarose gels........................................................... 20 2.2.1.12 Cloning strategy – EGFP expression vector ............................................ 20 2.2.1.13 Cloning strategy – vinculin full length and tail domain constructs.......... 21
2.2.2 Cell biology methods ....................................................................................... 22 2.2.2.1 Cell culture ................................................................................................... 22 2.2.2.2 Transient transfection of cells ...................................................................... 22 2.2.2.3 Culturing of cells on the cover slips for microscopic analysis..................... 23 2.2.2.4 Fixation and permeabilization...................................................................... 23 2.2.2.5 Immunolabelling, fluorescence microscopy and image processing............. 23 2.2.2.6 Determination of the spreading area ............................................................ 24 2.2.2.7 Fluorescence recovery after photo-bleaching (FRAP) measurements ......... 24 2.2.2.8 Electron microscopy measurements............................................................. 25
2.2.3 Biochemical and biophysical methods............................................................. 26 2.2.3.1 Bead coating................................................................................................. 26 2.2.3.2 Peptide synthesis .......................................................................................... 26 2.2.3.3 Lipid vesicle preparation.............................................................................. 26 2.2.3.4 Differential scanning calorimetry (DSC) measurements ............................. 27 2.2.3.5 Circular dichroism (CD) spectroscopy measurements................................. 29 2.2.3.6 Solid state NMR........................................................................................... 30 2.2.3.7 Cell lysis....................................................................................................... 32 2.2.3.8 Protein concentration determination according to Bradford ........................ 32 2.2.3.9 Western blot analysis ................................................................................... 32
2.2.4 Computational methods.................................................................................... 33
Table of contents
2.2.4.1 Molecular dynamics (MD) simulations........................................................ 33 2.2.4.2 Analysis of molecular dynamics simulations............................................... 34 2.2.4.3 Cluster analysis ............................................................................................ 35 2.2.4.4 Visualization of calculated structures .......................................................... 35
2.2.5 Cell mechanical methods ................................................................................. 35 2.2.5.1 Magnetic tweezer measurements ................................................................. 35 2.2.5.2 2D-traction microscopy measurements........................................................ 37
3 Results ............................................................................................................................. 38 I. In vitro ...................................................................................................................... 38 3.1 Differential scanning calorimetry (DSC) measurements ......................................... 38
3.1.1 DSC measurements of the C-terminal arm peptide.......................................... 38 3.1.2 DSC measurements of the mutated C-terminal arm......................................... 39
3.2 Molecular dynamics simulations.............................................................................. 40 3.2.1 MD Simulations of the C-terminal arm............................................................ 40 3.2.2 MD Simulations of the vinculin-tail................................................................. 46
3.3 CD-spectroscopy measurements of the C-terminal arm .......................................... 47 3.4 Solid state NMR measurements of the C-terminal arm ........................................... 49 II. In vivo ....................................................................................................................... 50 3.5 Cloning and expression of the different vinculin constructs.................................... 50
3.5.1 Cloning and expression of EGFP linked vinculin and vinculinΔC.................. 50 3.5.2 Cloning and expression of EGFP-linked vinculin-tail and vinculin-tailΔC..... 52 3.5.3 Mutagenesis of vinculin´s src dependent phosphorylation site Y1065F ......... 52
3.6 Magnetic tweezer measurements ............................................................................. 53 3.6.1 Magnetic tweezer measurements of MEF-resc and MEF-vinΔC cells ............ 54 3.6.2 Creep measurements of MEF-vtail and MEF-vtailΔC cells ............................ 58
3.7 Determination of the spreading area ........................................................................ 63 3.8 Determination of the FA per cell.............................................................................. 63 3.9 FRAP measurements ................................................................................................ 65 3.10 2D-traction microscopy measurements.................................................................... 66
3.10.1 Strain energy measurements of MEF-vinΔC ................................................... 66 3.10.2 Strain energy measurements of different vinculin mutants.............................. 67
4 Discussion ........................................................................................................................ 70 4.1 Vinculin´s lipid anchor revealed membrane insertion potential .............................. 71 4.2 Vinculin´s lipid anchor is involved in beta-sheet formation .................................... 72 4.3 Vinculin´s lipid anchor regulates cell stiffness ........................................................ 74 4.4 The lipid anchor affects traction generation via phosphotyrosine 1065 .................. 76 4.5 The Model ................................................................................................................ 78 4.6 Outlook..................................................................................................................... 79
5 References ....................................................................................................................... 81 6 Appendix ......................................................................................................................... 88
Abbreviations
A Alanine ADF Actin depolymerization factor AFM Atomic force microscope Amp Ampicillin AmpR Ampicillin-Resistenz Arp2/3 Actin related protein 2/3 ATP Adenosine-triphosphate ATPase Adenosine-triphosphate phosphatase BDM 2,3-Butanedione monoxime Bp Base pair BP320 Blocking peptide 320 BSA Bovine serum albumin C Cysteine Ca2+ Calcium CapZ Capping protein Z CD Circular dichroism cDNA complementary DNA CFP Cyane fluorescent protein CM Calcium/ Manganese CO2 Carbon dioxide CPU Central processing unit D Aspartate dATP Deoxyadenosine-triphosphat dCTP Deoxycytosine-triphosphat dGTP Deoxyguanosine-triphosphat Dia1 Diaphanous 1 peptide DMEM Dulbeccos modified eagle medium DMPC Dimyristoyl-L-α-phosphatidylcholine DMPG Dimyristoyl-L-α-phosphatidylglycerol DMSO Dimethyl sulfoxid DNA Deoxyribonucleic acid dNTP Deoxynucleotide-triphosphat DSC Differential scanning calorimetry DSSP Dictionary of secondary structure prediction DTT Dithiothreitol dTTP Deoxythymidine-triphosphate E Glutamate ECL-reagent Enhanced chemiluminescence reagent ECM Extracellular matrix EDTA Ethylenediaminetetraacetic acid EGFP Enhanced green fluorescent protein ERK-protein Extracellular signal regulated protein kinase F Phenylalanine FA Focal adhesion FAK Focal adhesion kinase FCS Fetal calf serum FERM F for Band 4.1, E for Ezrin, R for Radixin, M for Moesin FRAP Fluorescence recovery after photobleaching FRET Fluorescence resonance energy transfer Fs Femto-second Fx Focal complex
Abbreviations
g Gramm G Glycine Gd2+ Gadolinium GHz Giga-Herz GROMACS Groningen Machine for Chemical Simulations GTP Guanosine-triphosphate GTPase Guanosine-triphosphate phosphatase h Hour H Histidine HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid HMW marker High molecular weight marker HPC High performance computer I Isoleucine ILK Integrin linked kinase IRSp53 Insulin receptor substrate protein 53 K Lysine Kan Kanamycine kb Kilo-base Kcal Kilo-calories kDa Kilo-dalton kJ Kilo-joule kPa Kilo-pascal KT5926 A protein kinase inhibitor l Litre L Leucine LB Lura-Bertoni-medium LRW London resign white M Methionine mA milli-Ampere MCS Multiple Cloning site MD Molecular dynamics MEF Mouse embryonic fibroblast MgCl2 Magnesium-chloride MHZ Mega-Herz min Minute ML-7 Myosin light chain kinase inhibitor No. 7 MLV Multi-lamellar vesicles mM milli-molar N Asparagine Na+ Sodium NaCl Sodium-chlorid nm Nano-meter NMR Nuclear magnetic resonance ns Nano-seconds OD Optical density P Proline p.A. Per analysis p21 Cyclin-dependent kinase inhibitor 1A or CDKN1A p53 Protein or tumor protein of 53 kDa p95PKL Paxillin kinase linker PAA Polyacrylamide
Abbreviations
PAGE Polyacrylamide gel electrophoresis PAK p21 activated protein kinase PBS Phosphor buffered saline PC Phosphatidyl-choline PCR Polymerase chain reaction PDB Protein data base PEG Polyethylene glycol Pfu DNA polymerase from Pyrococcus furiosus pH Potential hydrogenii PH Pleckstrin homologue PI Isoelectric point PIP Phosphatidyl-inositol-phosphate PIP2 Phosphatidyl-inositol 4,5-bisphosphate PMSF Phenylmethanesulphonylfluoride POPC 1-Palmitoyl-2-Oleoyl-sn-Glycero-3-Phosphocholine POPG 1-Palmitoyl-2-Oleoyl-sn-Glycero-3-Phosphoglycerol ppm Parts per million ps Pico-second Q Glutamine R Arginine RAM Random access memory ROCK Rho associated kinase rpm Rotation per minute RPTP Receptor tyrosine phosphatases RT Room temperature s Second S Serine SCAR Suppressor of cAMP receptor SDS Sodiumdodecylsulfat SUV Small unilamellar vesicles T Threonine TAE Tris-Acetate buffer with EDTA Taq DNA polymerase from thermus aquaticus TBS-T Tris-buffered-saline with Tween TE Tris/ EDTA TEM Transmission electron microscope TEMED N,N,N’,N’, Tetraethylendiamin Tet Tetracycline TM Tris/ Magnesiumacetat Tris Tris-(hydroxymethyl)-Aminomethan TRITC Tetramethyl Rhodamine Iso-Thiocyanate V Valine Vinculin-CT EGFP-linked vinculin; RK1060/61 mutated to Q Vinculin-H3 EGFP-linked vinculin; K952, K956, R963, R966 mutated to Q Vinculin-LD Vinculin-CT + Vinculin-H3 VinculinY1065F EGFP-linked vinculin Y at position 1065 mutated to F VinculinΔC EGFP-linked vinculin (residues 1-1052) w.o. the lipid anchor Vol. Volumen Vtail EGFP-linked vinculin-tail (residues 858–1066) VtailΔC EGFP-linked vinculin-tail (residues 858-1052) w.o. the lipid
anchor
Abbreviations
W Tryptophane WASP Wiskott aldrich syndrome protein WAVE Wiskott aldrich Verprolin homologue protein Y Tyrosine YFP Yellow fluorescent protein
Zusammenfassung
Zusammenfassung Die extrazelluläre Matrix (ECM) beeinflusst und kontrolliert die Adhäsion sowie die
Migration von Zellen. Der fokale Adhäsionskontakt (FA) verbindet intrazellulär das Aktin-
Zellskelett über den transmembranen Integrin-Rezeptor mit Komponenten der ECM, wie
Kollagen und Fibronektin (FN). Der Auf- und Umbau dieser Adhäsionskontakte ist unter
anderem nur durch die Wechselwirkung von einigen fokalen Proteinen mit der Zellmembran
möglich. Das Vinkulin-Protein, welches in eine Kopf- (95 kDa) und eine Schwanzgruppe (30
kDa) unterteilt werden kann, zeigt solche Membran-bindende Strukturen. Es konnte
experimentell bestätig werden, dass Helix 3 (Aminosäuren 935-978) sowie die letzten 15
Aminosäuren des C-terminus (Aminosäuren 1052-1066; Lipid Anker) der Vinkulin-Schwanz
Gruppe mit Lipid-Vesikeln interagieren. Sogenannte Pull-down Experimente mit
Lipidvesikeln, die mit der Vinkulin-Schwanz gruppe ohne den Lipid Anker (vtailΔC)
inkubiert wurden, zeigten im Gegensatz zu den Versuchen mit dem gesamten Vinkulin-
Schwanz (vtail) keine Interaktion mit der artifiziellen Membran. Ob allerdings der Lipid-
Anker tatsächlich direkt an der Membran Wechselwirkung beteiligt ist, wurde im Rahmen
dieser Experimente nicht geklärt.
In dieser Arbeit wurde mittels „Differential scanning Calorimetry“ (DSC) versucht, die Frage
der direkten Beteiligung des Lipid-Ankers (Aminosäuren 1052-1066) an der
Membranbindung von Vinkulin zu klären. Diese Messungen haben gezeigt, dass der C-
terminale Arm von Vinkulin (Aminosäuren 1045-1066) mit dem hydrophoben Bereich der
Lipidvesikel in Kontakt tritt. Molekulardynamische Simulationen und Circular Dichroismus
(CD) Messungen lassen vermuten, dass der Lipid-Anker eine für Lipid-Interaktionen günstige
anti-parallele beta-Faltblatt Konformation einnimmt. Nuclear magnetic resonance (NMR)-
Messungen in Anwesenheit von POPC/ POPG Membranen bestätigten dies.
Weiterhin ist bekannt, dass Zellen die Vinkulin exprimieren welches nicht mit Membranen
wechselwirken kann, eine verminderte FA-Umbaurate aufweisen. Dies wirkt sich negativ auf
die Adhäsion und Migration der jeweiligen Zellen aus. Basierend auf diesen Ergebnissen
wurden im Rahmen dieser Arbeit in zusätzlichen in vivo Experimenten der Einfluss des Lipid-
Ankers auf die mechanischen Eigenschaften der Zellen getestet. Zu diesem Zweck wurden
MEF-Vinkulin(-/-) Zellen mit Vinkulin dessen Lipid-Anker fehlte (MEF-vinΔC)
retransfiziert. Diese MEF-vinΔC Zellen wurden mit Fibronectin beschichteten
paramagnetischen „beads“ inkubiert und mit einer „magnetischen Nadel“ untersucht. Im
Vergleich zu wildtyp (MEF-wt) und rescue (MEF-resc) Zellen verhielten sich die MEF-
vinΔC Zellen weniger steif. Gleichzeitig rissen während der Messung von diesen Zellen mehr
Zusammenfassung
„beads“ ab. Dies ist ein Indiz für die verminderte Anbindung an ECM beschichteten
Oberflächen von MEF-vinΔC Zellen. 2D-traction microscopy Experimente zeigten weiter,
dass diese MEF-vinΔC Zellen im Vergleich zu MEF-wt bzw. MEF-resc Zellen während der
Adhäsion geringere Kräfte und weniger fokale Kontakte ausbilden. Dies wird auf die
verminderte Actin-myosin Aktivität zurückgeführt. Messungen der Kraftentwicklung von
Zellen, die weitere Lipidbindedefiziente Varianten von Vinkulin exprimierten, zeigten, dass
lediglich der Knock-out des Lipid-Ankers von Vinkulin zu einer Reduktion der Kräfte führt.
Der Knock-out der src-Phosphorylierungstelle im Lipid-Anker (Y1065F) führt zu ähnlich
verringerten Kräften. Diese lässt den Schluss zu, dass die Zellmechanik über die
Membranbindung des Lipid-Ankers, sowie der src-abhängige Phosphorylierung von Vinkulin
beeinflusst wird.
Summary
Summary The contact of the cell to the extracellular matrix components such as collagen and fibronectin
is important for cell adhesion and migration. Most of the cell matrix contacts are linked to the
actin filaments by the focal adhesions (FA). At the FA, heterodimeric transmembrane integrin
receptors link the ECM to the cytoskeleton via adaptor proteins that are part of a sub
membrane plaque. A variety of those adaptor proteins have been shown to associate with, and
in some cases insert into, lipid bilayers. The focal adhesion protein vinculin (1066 residues),
which can be separated into a 95 kDa head and a 30 kDa tail domain, shows such lipid
binding sites. However, the function of vinculin´s lipid binding is still an enigma. Two
regions on the 30 kDa tail domain have been experimentally identified as candidates for lipid-
binding: Helix 3 (residues 935–978) and the lipid anchor (residues 1052–1066). The alteration
of helix 3 (residues 944-978) and the unstructured C-terminal arm (residues 1052-1066, so-
called lipid anchor) resulted in impaired lipid vesicle interaction of the vinculin-tail. Pull-
down assays with artificial lipid membranes revealed that in contrast to vinculin-tail (vt), a
variant lacking the lipid anchor (VtΔC), does not interact with vesicles. To what extent the
last 15 residues are involved in lipid interaction was not determined.
In this study the lipid-binding ability of the lipid anchor of vinculin as well as the influence to
cell mechanical behavior were determined. Differential scanning calorimetry (DSC)
demonstrated that vinculin´s C-terminal arm, which includes the lipid anchor, is directly
involved in lipid binding. The peptide inserts into the lipid vesicle consisting of DMPC/
DMPG at various molar ratios. The secondary structure of the C-terminal arm was also
explored under different ionic conditions which represent nominal basic, neutral and acidic
pH´s using molecular dynamics simulations. The generated trajectories predicted an
antiparallel beta-sheet followed by an unstructured C-terminal end for the peptide
representing vinculin´s C-terminal arm under “basic” and “neutral” conditions. This
conformational behavior was investigated in more detail in the presence/absence of DMPC/
DMPG vesicles using CD-spectroscopy. The results suggest direct association of vinculin’s
lipid-binding region (residues 1052–1066) with membranes whilst forming a beta-sheet. To
determine the orientation of the lipid anchor during membrane interaction, solid state NMR
measurements were performed using vinculin´s C-terminal arm peptide. Those results imply
that in presence of POPC/ POPG vesicles, the beta-sheet inserts into the lipid membrane.
Furhermore, it was demonstrated that cells expressing vinculin without the lipid anchor
(vinΔC) showed a decreased focal adhesion turnover rate, which results in impaired cell
adhesion and migration. In additional in vivo experiments, the influence of the lipid anchor
Summary
region (residues 1052–1066) in terms of cell mechanical behavior was determined using
vinculin deficient mouse embryonic fibroblasts, retransfected with EGFP-linked vinculin
lacking the lipid anchor (MEF-vinΔC). Magnetic tweezer experiments revealed that MEF-
vinΔC cells, incubated with fibronectin coated paramagnetic beads, were less stiff and more
beads detached during these experiments compared to MEF-resc cells. Cells expressing
vinΔC formed fewer focal contacts as determined by confocal microscopy. 2D-traction
measurements showed that MEF-vinΔC cells generated less force compared to rescue cells.
Attenuated traction forces were also found in cells that expressed vinculin with point
mutations of the lipid anchor that either impaired lipid binding or prevented src-
phosphorylation at site Y1065. However, traction generation was not diminished in cells that
expressed vinculin with impaired lipid binding due to point mutations on helix 3. These
results show that both the lipid binding and the src-phosphorylation of vinculin's C-terminus
are important for cell mechanical behavior, but the lipid binding of helix 3 is not, suggesting
that both the lipid anchor and the src-phosphorylation of Y1065 affect cell mechanical
behavior.
Introduction
1
1 Introduction
1.1 The extracellular matrix determines the different tissues Most of the eucaryotic cells in multicellular organisms are organized as cooperative
assemblies called tissue, which in turn are associated in various combinations to form larger
functional units called organs [1]. In higher organisms such as vertebrates, the major tissue
components are nerves, muscles, blood, lymphoid, epithelial and connective (mesenchymal)
tissue. The cells in tissues are usually in contact with a complex network of extracellular
macromolecules referred to as extracellular matrix (ECM) which is mainly secreted by
fibroblasts. This matrix organizes the contact to neighbouring cells and provides a lattice for
cells to migrate. The matrix also determines the localized function of the tissue. To maintain
the tissue, the cells need mechanical strength, which is often provided by the extracellular
matrix. Connective- and epithelial tissues represent two extremes in which structural roles
played by the matrix and by cell-cell contacts are radically different (figure 1-1). In
connective tissues, the extracellular matrix is abundant and cells are sparsely distributed in it.
The matrix is full of fibrous polymers, such as collagen. Cells as well as the matrix bear most
of the mechanical perturbations to which the tissue is subjected. The cells such as fibroblasts
are attached to the ECM components and exert forces, whilst direct interactions of the
embedded cells are relatively unimportant. In contrast, epithelial tissues are built of tightly
bound cells which form layers. The ECM is here sparsely distributed and consists mainly of a
thin mat called basal lamina that can be found under each cell layer.
Figure 1-1: Simplified cartoon of a cross section through part of the intestine [1]. Here, the cells themselves develop and deal with mechanical stresses by protein filaments
such as the cytoskeleton. The components criss-cross the cytoplasm of each epithelial cell to
Introduction
2
transmit any mechanical perturbations from one cell to the other. The cytoskeleton is directly
or indirectly linked to transmembrane proteins in the plasma membrane where specialized
junctions connect the surface of adjacent cells with the underlying basal lamina. Epithelial
cell sheets line all the cavities and free surfaces, and the specialized junctions between the
cells enable these sheets to form barriers between solutions and cells from one body
compartment to another. Epithelial sheets nearly always rest on a supporting bed of
connective tissue, which may attach them to other tissue types such as muscle [1].
The formation and maintenance of tissue is dependent on the ability to resist surface tension
and shear stress [2]. Mechanical loads exerted at the macroscale affect conformational
changes in connective tissue and ECM. This results in production of stress at the microscale
level that produces global structural rearrangements of the individual molecular components
such as the ECM and the interconnected cytoskeleton of adherent cells. Figure 1-2 shows that
tension-dependent changes in the orientation of collagen bundles, cytoskeletal and nuclear
rearrangements within adherent fibroblasts, as well as distortion of interconnected basement
membrane scaffolds. Those in turn produce similar cytoskeletal and nuclear rearrangements
within associated endothelial cells.
Figure 1-2: Cellular mechano responsiveness and physical connectivity between ECM, cells and the cytoskeletal network. The shear stress and the cells´ own cytoskeletal movement results in changes of the collagen fibers conformation [2].
At the same time, fluid shear stresses applied at the apical surface of certain endothelial and
epithelial cells can also influence the cytoskeletal structure by deflecting the primary cilium
found on the surface of many cells, and by exerting tension on the plasma membrane and
apical cell-cell junctions. Because these shear stresses are also channelled through the
cytoskeleton to basal cell-ECM adhesions, they can alter the structure and function of
underlying connective tissue as well. The entire cellular response to stress may therefore vary
depending on the structural integrity and organization of the whole cytoskeleton-cell-ECM
Introduction
3
lattice, as well as the tension in the network prior to load application. For instance, in the lung
the residual filling pressure that remains after expiration is responsible for tensing and
stiffening the ECMs (basement membranes, collagen fibers, elastin bundles) that surround
each alveolus, and for resisting surface tension forces acting on the epithelium. This force
balance stabilizes the alveoli in an open form [3]. Lung expiration and inspiration influence
this force balance and produce complex micromechanical responses in the lung parenchyma,
including lengthening and shortening (and tension and compression) of alveolar walls
depending on the direction of the applied stress [2]. This is accompanied by extension and
linearization of some collagen fibers on inspiration, as well as buckling of the same fibers on
expiration. Breathing also causes the lateral intercellular spaces between epithelial cells to
reversibly shrink and expand without compromising the structural integrity of the tissue.
Those form of reversible mechanical deformation within surrounding alveolar cells by
altering the local concentration of soluble ligands for epidermal growth factor receptors [4].
1.2 Mechanical perturbations regulate cell survival and tissue formation Throughout their life, cells participate in numerous physical interactions with their
neighbourhood. Endothelial cells sense physical perturbations such as shear stress, produced
by the blood flow [2; 5-6]. In kidney epithelial cells, fluid shear stress produces Ca2+ influx by
deflecting the primary cilium, which acts like a long, microscopic, vertical lever arm on the
apical surface of the cells [2]. In the lung the residual filling pressure that remains after
expiration is responsible for tensing and stiffening the tissue. Cells embedded into the
muscular tissue are dealing with mechanical pressure, generated by their one contractile
machinery such as the actomyosin filaments [5-6]. At cell contact sites, the extracellular
domains of the transmembrane adhesion molecules interact with partners localized on the
surface of the adjacent cells or in the extracellular matrix, whilst the intracellular parts are
attached to components of the cytoskeleton. The connections provide anchor points to mediate
the mechanical stability and integrity of the cells. Furthermore, cell-cell and cell-matrix
contacts are important for the formation and preservation of tissues, cell survival and
proliferation. In order to form and maintain such tissues, eucaryotic cells have elaborated
systems that enable the cell to interpret and amplify signals from other cells. The signalling
systems includes cell surface and intracellular receptor proteins, protein kinases, protein
phosphatases, GTP-binding proteins and many other proteins with which those proteins
interact. This classical process so called signal transduction refers to any process by which a
cell converts one kind of a stimulus into biochemical signal [1]. Additional, an increasing
Introduction
4
body of evidence suggests that besides the classical biochemical signals such as cytokines and
hormones, force dependent mechanical interactions play a central role in regulating cell
behavior [7]. Cell adhesion to other cells or extracellular matrix components (ECM) such as
collagen, fibronectin and laminin trigger biochemical or biomechanical signals inside the cells
that initiate cell survival, proliferation or cell migration by regulating protein synthesis and
cytoskeletal reorganization [2; 7-9]. Most of the cell-cell and cell matrix contacts are linked to the intermediate- or actin filaments.
These transmembrane driven connections are summarized and described in figure 1-3.
Tight junctions are structures between two endothelial or epithelial cells and prevent the
leakage of water-soluble molecules into the tissue. They serve as semi-permeable diffusion
barrier between two cells by separating distinguished areas. These structures are Ca2+
dependent built. Chelating Ca2+ ions lead to tight junction dissolution. In order to transport
high amounts of amino acids and glucose, the tight junctions can be opened. Integral
transmembrane proteins such as claudin, and occluding link apically two neighbouring cell.
These peptide structures are intracellular linked to the actin cytoskeleton by adaptor peptides
form the zonula occludens (ZO) family [10].
Adherens junctions are adhesion belts, commonly found between epithelial cells in the
small intestine. The belt-like junction encircles each of the interacting cells. Its most obvious
feature is a contractile bundle of actin filaments running along the cytoplasmic surface of the
junctional plasma membrane. In this intercellular connection, the actin filaments are joined
from cell to cell by transmembrane adhesion proteins called cadherins. The cadherins form
homodimers in the plasma membrane of each interacting cell. The extracellular domain of one
cadherin dimer binds to the extracellular domain of an identical cadherin dimer on the
adjacent cell. The intracellular tails of the cadherins bind to anchor proteins that tie them to
actin filaments. These anchor proteins include α-catenin, β-catenin, γ-catenin (also called
plakoglobin), α-actinin, and vinculin. [9-10].
Desmosomes are button-like points of intercellular structures that link mainly epithelial and
heart muscle cells. The desmoglein and desmocollin glycoprotein receptors (also cadherin
superfamily) are indirectly connected with intermediate filaments such as keratin and desmin.
Desmoplactine and plakoglobine built the junction between the receptor and keratin-
filaments. Extracellular the receptors are in contact with cadherin receptors of the same
family, exposed by the neighbouring cell [11]. Those structures prevent the leakage of body
fluids into the loosened epithelium.
Introduction
5
Hemidesmosomes or half desmosomes are morphologically similar to Desmosomes.
Instead of joining adjacent epithelial cell membranes, they connect the basal surface of those
cells via integrins to extracellular matrices (ECM) consisting of laminin – a specialized mat of
ECM at the interface between the epithelium and the connective tissue. Intracellular, the
receptors are linked to the intermediate filaments by plectin and BP230 adaptor peptides [12].
Focal adhesions are the structural junction between the ECM via integrins with the
cytoskeleton. Structural proteins such as talin, alpha actinin and vinculin built the linkage
between the receptors and the actin filaments. Focal adhesions serve as the mechanical
linkages to the ECM, and as a biochemical signalling hub that concentrate and directs
numerous signaling proteins. In sessile cells, focal adhesions are quite stable whilst in moving
cells their stability is diminished: this is because in motile cells, focal adhesions are being
constantly assembled and disassembled as the cell establishes new contacts at the leading
edge, and breaks old contacts at the trailing edge of the cell. One example of their important
role is in the immune system, in which macrophages migrate along the connective
endothelium following cellular signals to damaged tissue [2; 5-8].
Figure 1-3: Schematic of cell contacts visualized in epithelial cells [1].
The most prominent mechanosensory devices that can detect forces and respond to them are
adhesion sites to neighbouring cells and to the ECM. However, the function of such “force
receptors” is still poorly understood and characterized in comparison to the classical
biochemical receptors [2, 5-8]. The most prominent and best characterized mechano-receptor
is the focal adhesion complex or focal contact. Ingber and co-workers showed in magnetic
Introduction
6
twisting experiments that the permanent movement of integrin ligand coated beads, attached
to cells, induces a force dependent “stiffening response” in those cells. During these
experiments, cells were incubated with ECM-coated magnetized beads. To the particles a
magnetic field was applied which in turn displaces the beads connected to the cells. Due to the
applied stress, the cells increased their resistance to mechanical deformation, suggesting that
the focal adhesions respond to applied forces [13].
1.3 The focal adhesion complex Structurally defined focal adhesion sites between cultured cells and ECM were described
about 30 years ago using interference reflection, fluorescence and electron microscopy [5; 7;
14-15]. The focal adhesions anchor the cells to the ECM and can be described as streak like
structures which are connected to actin stressfibers (figure 1-4A). Besides anchorage of cells
to the ECM, focal adhesion assembly is also involved in cell migration. Focal adhesion (FA)
complexes are highly dynamic structures composed of structural and signal transduction
proteins. As described in figure 1-3, heterodimeric transmembrane integrin receptors link the
ECM to the cytoskeleton via adaptor proteins that are part of a sub membrane plaque (figure
1-4B). The beta subunits of integrins play a direct role in the establishment of that connection.
The intracellular part of integrin exhibits two conservative NPxY and NPxF motives. With
those sites the receptor can interact with peptides that show phosphotyrosine binding sites
(PTB like modules).
Figure 1-4: Illustration of focal adhesions. A) Visualization of the focal adhesion complexes (green), stained with EGFP-linked vinculin in mouse embryonic fibroblast. The focal adhesions form streak like structures which are in contact with the actin stress-fibers (red). The bar represents 20 µm. B) A schematic diagram of a focal adhesion complex, modified according to Goldmann et al [16]. The integrin receptors link the ECM to the actin cytoskeleton via the sub membrane plaque. This plaque is composed of structural and signalling proteins which sense mechanical perturbations and translate them into biochemical signals. Talin belongs to that group and connects the integrins directly to the actin cytoskeleton [17].
Other actin related peptides such as alpha actinin and filamin also link the transmembrane
Introduction
7
receptor to the cytoskeleton [7; 18]. In addition to these direct linkers, there are structural
components such as vinculin, actin related protein (Arp2/3) or paxillin that reinforce the
linkage to the cytoskeleton and/or recruit signal transduction proteins to the focal adhesions
[19- 21].
Peptides such as small GTPases (Rho, Rho associated kinase (ROCK), Rac etc.),
phosphotyrosine kinases (focal adhesion kinase (FAK), Src, Fyn), Phosphotyrosine
phosphatase, serin threonine kinases (integrin linked kinase (ILK) or p21 activated kinase
(PAK)) and phosphatidyl-inositol kinases are also localized in the focal adhesions [5; 7; 22].
Altogether, these molecular conductors regulate FA formation by integrating the mechanical
perturbations into biochemical signals such as phosphorylation [5]. This implies
reorganization of the actin cytoskeleton and conversion of physical signals, such as contractile
forces or extracellular matrix perturbations into chemical signals that results in peptide
synthesis. In particular the FAK/Src pathway seems to be involved in the regulation of FA
turnover [18]. FAK (-/-) fibroblasts are deficient in both force–induced FA growth and cell
response to substrate rigidity [23]. Recently receptor tyrosine phosphatases alpha (RPTP-
alpha) which is involved in activation of src family kinases is necessary for FA reinforcement
of αv β3 integrin cytoskeleton connections [24].
Tensions are essential for FA maintenance and development [25-27]. This is reached by using
the contractile actomyosin apparatus. The actin filaments are highly dynamic structures
distributed over the whole cell body. Those filaments are in contact with the myosin motor-
proteins. Activating myosin by phosphorylation causes filament sliding and pre-stress
generation. The inhibition of the myosin II-driven contractility causes a reduction in the
formation of new focal adhesions, and the destabilization of the existing ones [25-27]. The
reduction of myosin II-driven tension by substances such as ML-7, BDM, KT5926 or by the
overexpression of peptides like caldesmon (inhibits actin-dependent myosin II ATPases
activity), brings about a rapid decrease of the focal adhesion size. The block of contractility
leads to complete dissolution of focal adhesions [25-27].
The formation and stability of force-dependent focal adhesion are not only due to the cell’s
own contractile system. The stimulus can also be applied from outside [28-29]. Indeed,
external force applied to focal adhesions can effectively substitute cell-generated forces even
in chemically relaxed cells. It was demonstrated that focal adhesions may be stimulated to
grow even in relaxed cells treated by BDM or transfected with caldesmon by pulling on single
cells using a micropipette [29]. Mechanical force can also be applied to micron-sized beads,
coated with fibronectin or other extracellular matrix proteins attached to the dorsal cell
Introduction
8
surface. Without application of external force, these beads move centripetally, driven by
retrograde actin flow [30]. However, trapping of a bead with laser tweezers directs the force
produced by the flow to the integrin-mediated contact at the interface between the bead and
the cell surface. This force was shown to strengthen the transmembrane link between the bead
and the actin cytoskeleton, which further increases the flow-driven force applied to the bead
and eventually allows it to escape from the trap [31]. This process, termed reinforcement,
seems analogous to the process of focal adhesion assembly and similarly includes recruitment
of the new vinculin molecules into the adhesion plaque [32].
An additional important factor is the mechanical nature of the underlying substrate. When
cells are attached to a soft flexible substrate, which can be easily deformed, the tension acting
on the adhesion plaques may be smaller than the force needed to sustain the adhesion site and
the attached stress-fibers. Consequently, the typical dimensions of focal adhesions formed
with such substrates are considerably smaller than those formed following attachment to a
rigid surface [33]. This ability to discriminate between soft and rigid substrate enables cells to
become oriented [34].
1.4 Focal adhesion formation Focal adhesions (FAs) evolve from “dot-like” structures, smaller than 1 µm in diameter
adhesion sites, termed focal complexes (FXs)[5]. They are nascent integrin- mediated
adhesions in protrusive lamellipodia developed in a Rac dependent manner (figure 1-5) [18;
35-36]. It was demonstrated that the Rac-GTPase gets recruited to the ECM attached
lammelipodia [37] and influences focal complex (FX) formation in three possible ways.
GTPases as Rac are membrane linked peptides which get activated by binding the nucleotide
guanosine- tri-phosphate (GTP). The autocatalytic transformation of GTP to Guanosine-di-
phosphate (GDP) deactivates this peptide. Rac activates the phosphatidyl inositol phosphate
kinase (PIP 5-kinase) which generates phosphatidyl inositol 4-5 bisphosphate (PIP2).
The membrane-associated inositol variant has the ability to uncap existing actin filaments
near the membrane and activate FA molecules such as talin and vinculin [38]. It is believed
that talin is involved in sequestering/ activating integrins. Furthermore, the GTPase Rac
initiate the WAVE/SCAR (peptides from the Wiskott Aldrich syndrome protein or WASP
family) peptide complex formation via the insulin receptor substrate p53 (IRSp53) that in turn
activates the actin nucleator Arp2/3 [39-40]. The serine-threonine kinase PAK is also
activated by Rac [41]. PAK promotes actin polymerization via LIM-Kinase, a kinase which is
deactivating ADF/Cofilin, an actin depolymerization factor. This prevents actin filament
Introduction
9
dissolution in the lamellipodia. Further, PAK gets localized to the membrane after the cell
attaches to ECM molecules. Here it associates with integrins, possibly via paxillin and the
paxillin kinase linker p95PKL [42].
Figure 1-5: Schematics of the Rac dependent focal complex formation (A) and the Rho dependent maturation to focal adhesions (B). Rho signalling is responsible for generating the appropriate tension generation which is necessary for recruiting other actin linked integrins for focal adhesion maturation.
The transformation from FX to FA is force dependent. Experiments using EGFP β3 revealed a
myosin-dependent recruitment of integrin αv β3 during FA maturation [43-44]. In addition,
myosin-driven force extracts molecular complexes consisting αv β1 integrins and tensin from
the focal contact [45-46]. These structures known as fibrillar adhesions move centripetally
and transmit the myosin driven forces to fibronectin molecules [18]. Integrin signalling also
activates Rho kinase, but the mechanism is not clear [5]. Rho coordinates the force-dependent
growth and maturation of focal complexes to focal adhesions [35-36]. It is known that the rho
targets ROCK (Rho associated kinase) and the formin homolog peptide Dia 1 (Diaphanous 1
peptide) are necessary for FA formation [5; 18]). ROCK is known to activate myosin-II by the
phosphorylation of the myosin light chain (MLC) [47]. Inhibition of ROCK function by
chemical inhibitors or dominant negative mutants prevents the formation of FA contacts [29;
48], but not of focal complexes. Dia 1 is also involved in actin cytoskeleton regulation.
Experiments suggest Dia 1 promotes actin polymerization by targeting profilin and Arp2/3
activation. It was also demonstrated that Dia 1 targeting microtubules to the FA [49].
Introduction
10
The focal adhesion mechanosensor is probably not a unique molecular machine but a
prototypic device, which can have analogs in other cellular systems such as adherens junction
complexes and the Z-disc-complex in striated muscle. Focal adhesions have the features to act
as mechanosensors, probe the physical environment and activate certain signal cascades. This
capacity to convert physical perturbations into biochemical signals is based on the existence
of structural and signalling components in the adhesion plaque. However, some of the basic
functions of the mechanosensing apparatus are still unknown. It is still an enigma which
molecules in the focal adhesions sense and transmit the force [5-6; 18]. There are several
promising candidates to act as mechanosensor such as vinculin [20]. For example, blocking
the Ca2+ signalling in fibroblasts with Gd2+, which regulates local traction forces, correlates
with the reduction of vinculin at the focal adhesions which resulted in impaired cell migration
[50]. Furthermore, fluid shear measurements with NIH 3T3 cells revealed that adhesion
strengthening is accompanied by a 300 % increase in vinculin and a 90 % increase in
recruitment of talin to adhesion structures, whereas integrin levels were unchanged [51].
1.5 Vinculin, a key player of the focal contact The 116 kDa vinculin with its 95 kDa head- and 30 kDa tail domain is kinetically one of the
first proteins in FA formation [16]. As demonstrated in figure 1-6, it is available in two states,
the open and the closed conformation [20]. In the closed or autoinhibited state the vinculin-
head keeps its tail pincer-like in place. Activating vinculin frees the binding sites on the tail
for paxillin, actin and phospholipids [19-20]. Several experiments revealed that vinculin is
only in the open state localized at the FA [52-53]. For these experiments, a vinculin FRET
construct with a yellow fluorescent protein (YFP) label in the prolin rich domain and a cyan
fluorophor (CFP) at the C-terminus was developed. The FRET ratio of these constructs was
decreased in the FAs in comparison to the cytosol. This suggests that the recruitment of
vinculin to the focal adhesions is linked with a conformational shift which displaces the
vinculin-head from the tail [52]. Additional studies with this FRET construct revealed that
talin together with a second binding partner (actin) have the ability to displace the vinculin-
head from the tail [54]. Talin interacts with the vinculin-head region (residues 1-258) that
lowers the affinity between the head and tail domain [55-57]. The microinjection of these
vinculin binding sites specifically targets vinculin in cells, disrupting its interactions with talin
and disassembling focal adhesions [57]. Besides talin, alpha-actinin shows also vinculin
activation. Alpha-actinin binds at the same position at the vinculin-head (residue 1-258) as
talin [57-58]. It was also reported that acidic phospholipids such as phosphatidyl-inostiol 4-5
Introduction
11
bisphosphat (PIP2) can cause a similar conformational shift and activate vinculin [59-61].
Experiments with acidic phospholipids incubated with the vinculin-head and the vinculin-tail
domain demonstrated that PIP2 vesicles have the ability to displace the head from the tail [59].
Additional in vitro experiments could demonstrate that the 30kDa vinculin-tail binds to
negatively charged lipid vesicles [60]. Johnson and co-workers incubated the 30kDa tail and
the 95kDa head fragment separately with phosphatidyl-inositol (PI) vesicles. Only the
vinculin-tail could be found together with the PI-vesicles under physiological ionic
conditions. The vinculin-head region showed no lipid interaction potential.
Cells deficient of vinculin are still able to form focal contacts but spread poorly on ECM-
coated surfaces [62-65]. They are more motile and less adhesive than wildtype cells [62; 64].
Furthermore vinculin (-/-) cells showed a higher level on tyrosine phosphorylated proteins
localized in the focal adhesions [64]. Stable retransformation of the vinculin-tail in F9 mouse
carcinoma vinculin (-/-) cells lead to improved adhesion and decreased cell motility compared
to F9 vinculin (-/-) cells [66]. Reintroduction of the complete vinculin molecule rescued the
wildtype phenotype.
Figure 1-6: Schematic of the vinculin molecule in the closed/auto inhibited (a) or the open/active (b) conformation according to Critchley [19]. About 8 % of intracellular vinculin are necessary to recover the wildtype phenotype [66].
Magnetic twisting, magnetic tweezer and atomic force microscope (AFM) measurements of
mouse F9 vinculin (-/-) cells showed softer behavior of vinculin knockout cells compared to
Introduction
12
F9 wildtype cells, suggesting that vinculin acts as an intracellular mechano-coupler [63; 65;
67-69]. It is also believed that phosphorylation of vinculin is important for the mechano-
coupling function of vinculin [70-72]. Removing the phosphorylation site on position Y822
causes an up-regulation of p-ERK which resultes in the reduction of cell migration [70].
Furthermore, c-src-dependent vinculin phosphorylation on position Y100 and Y1065 affects
cell spreading and migration, indicating that the phosphorylation of vinculin stabilizes the
active or open conformation [71]. It was further demonstrated, that phospholipid interaction
of vinculin modifies its c-src- dependent phosphorylation [73-74]. In the presence of acidic
phospholipids vesicles such as phosphatidyl-inositol (PI), the phosphorylated fraction of
vinculin was elevated more than 10 fold compared to vinculin in absence of lipid vesicles
[73].
1.5.1 Lipid binding of vinculin
Since it is known that signal transduction, vesicle trafficking, retroviral assembly, and other
central biological processes involve the directed binding of proteins to membranes,
researchers started to try to elucidate their mechanisms and functions [38]. Soluble proteins
may associate with membranes through well-defined structural domains such as amphipathic
helices and/or unstructured motifs that interact through non-specific electrostatic and apolar
interactions. Post-translational modifications, such as myristylation or palmitoylation, may
also play critical roles in regulating membrane association of peptides. Many cytoskeleton-
associated proteins such as alpha-actinin, talin, Arp2/3, CapZ and vinculin interact, at least
transiently, with lipid membranes [38; 75].
Three regions on the 30 kDa tail domain of vinculin have been identified as candidates for
lipid-binding: residues 935–978, 1020–1040 and 1052–1066 (figure 1-7) [76-78]. Residues
935–978 and residues 1020–1040 were identified as amphiphatic alpha-helices by an
algorithm that recognized highly hydrophobic or amphiphatic amino acid segments with
alpha–helical character, whilst discriminating between surface-seeking and transmembrane
configurations [38; 76]. Johnson and co-workers experimentally confirmed the computational
prediction. They incubated several peptides of the vinculin-tail with radioactive-labeled
phosphatidylserine (PS) lipid vesicles. The fragments of residues 916–970 inserted into the
hydrophobic core of lipid vesicles. Crystal structure analysis revealed an alpha-helical
secondary structure for this lipid binding site [55; 78; 79]. The residues 1052–1066, which
also have lipid interaction potential, remained unstructured during crystallization. Differential
scanning calorimetry (DSC) together with circular dichroism (CD) spectroscopy
Introduction
13
measurements showed that this lipid anchor region of vinculin has the potential to insert into
the hydrophobic core of lipid vesicles consisting of dimyristoyl-L-α-phosphatidylcholin
(DMPC) and dimyristoyl-L-α-phosphatidylglycerol (DMPG) whilst forming a beta-sheet [80].
Pull-down assays with artificial lipid membranes demonstrated that the vinculin-tail variant,
lacking the last 15 amino acids (VtailΔC) was not able to interact with acidic
phosphatidylserine (PS) or phosphatidylinositol (PI) vesicles under physiological conditions
compared to full length vinculin-tail (Vtail) [78; 81].
Figure 1-7: The vinculin-tail with its three potential lipid binding sites (colored in gold), obtained from the crystal structure of chicken vinculin (PDB 1ST6) [55]. It was also shown that the lipid interaction of the vinculin-tail is mainly driven by electrostatic
interaction between the lipid vesicle and peptide surfaces. As demonstrated by additional pull-
down experiments, the mutation of six surface exposed basic residues on vinculin-tail to
glutamine (Q) in helix 3 (K952, K956, R963, R966) and the C-terminal lipid anchor
(RK1060/61) also significantly reduces the interaction potential with negatively charged
vesicles [82]. Furthermore, it was reported that cells transfected with a vinculin variant
including this 6 point mutations which cause lipid binding deficiency (vinculin-LD) showed a
reduced focal adhesion turnover rate and impaired cell adhesion on different extracellular
substrates such as collagen, laminin or fibronectin. The cell motility in vinculin-LD cells was
also decreased [82]. Cells expressing vinculin without the lipid anchor (vinculinΔC) showed
Introduction
14
the same reduced FA adhesion turnover rate and decreased cell motility [81]. These results
imply that only the membrane interaction of vinculin´s lipid anchor influences cell
mechanical behavior.
1.6 Aims Lipid binding of vinculin influences the focal adhesion turnover rate, which results in
impaired cell adhesion and migration [82-81]. Pull-down assays with artificial lipid
membranes consisting of phosphatidylserine (PS) or a mixture of 40 % phosphatidylinositol-
4,5-bisphosphate (PIP2) and 60 % phosphatidylcholine (PC) revealed that in contrast to Vtail,
a variant lacking the last 15 amino acids (VtailΔC), does not interact with vesicles of these
compositions [78; 81]. But whether the last 15 residues are directly involved in lipid
interaction was not determined. Crystal structure analysis together with CD spectroscopy
analysis revealed that the integrity of the helical bundle is ensured in both vinculin-tail
variants, even in the absence of the head domain [78; 81]. In this work, we want to clearify
the question wether the lipid anchor is directly or indirectly involved in lipid membrane
interaction. Furthermore the impact of that interaction due to cell mechanical behavior was
investigated. To determine the membrane interaction of the lipid anchor region, differential
scanning calorimetry (DSC) measurements were performed, using a peptide which represents
vinculin´s unstructured C-terminal arm. In additional molecular dynamics simulations we
calculated possible secondary structures for the peptide. CD-spectroscopy as well as NMR-
measurements were performed to verify the calculated secondary structure and calorimetric
characterization of the peptide. Furthermore, the influence of the lipid anchor due to cell
mechanical behavior was determined using a vinculin variant lacking the last 15 residues
(vinculinΔC). Transient with EGFP-labeled vinculin∆C transfected MEF vinculin(-/-) cells
were examined using a magnetic tweezer setup with force feedback control. Forces between
0.5 and 10 nN were applied in a log-step force protocol to investigate the stiffness of the cells
as well as the stability of the FN-integrin-actin linkage. The actomyosin- driven contractile
forces of MEF-vinΔC cells were obtained by 2D-traction microscopy measurements. Further,
the numbers of FA as well as the turnover rate of the different vinculin constructs were
determined.
Materials & Methods
15
2 Materials and Methods 2.1 Materials 2.1.1 Chemicals and enzymes
All chemicals were of per analysis (p.A) quality and purchased from Sigma (Deisenhofen),
Merck (Darmstadt), Fluka (Neu-Ulm), Difco (Hamburg) and Invitrogen (Karlsruhe). All
enzymes were from the following companies unless otherwise specified, New England
Biolabs (Bad Schwalbach), Stratagene (Heidelberg), Boehringer (Mannheim), Amersham
(Freiburg), Invitrogen (Karlsruhe), and Promega (Mannheim).
2.1.2 Cell culture medium and plastic ware
Cell culture media and additives such as fetal bovine serum (FBS) and the antibiotics
Penicillin and Streptomycine were purchased from Gibco (Karlsruhe) and Sigma
(Deisenhofen). Plastic ware was obtained from Corning (Kaiserslautern), Eppendorf
(Hamburg), Becton and Dickenson (BD-Bioscience, Heidelberg), Greiner (Frickenhausen)
and Nunc (Langenselbold).
2.1.3 Oligonucleotides
The Oligonucleotides employed in cloning and sequencing reactions were obtained from
MWG-Biotech (Ebersberg). The primers for sequencing cloning and mutagenesis are listed in
Table 2.1, and primers for cloning and mutagenesis are listed in Table 2.2. The restriction
sites present in the primers are specified.
Number Name Target sequence Sequence 5’ to 3’ Primer P1 Seq_pEGFP_for1 pEGFP, start in front of
EGFP TGGGCGGTAGGCGTGTA
Primer P2 Seq pEGFP rev1 pEGFP, start at the end of EGFP
CAGGTTCAGGGGAGGT
Primer P3 Seq_EGFP_rev pEGFP, start in the middle of EGFP
GAGCAGGATGGGCAC
Primer P4 Seq_pcDNA_rev pcDNA3.1/Hygro, behind MCS
GGTCAAGGAAGGCACG
Primer V1 Seq vinc rev1 mvinculin, start at 703bp-720bp
TTCAGCACTCATCTTTTC
Primer V2 Seq vinc for1 mvinculin, start at 551bp-568bp
ACCAGGAACACCGTGTG
Primer V3 Seq vinc rev2 mvinculin, start at 1953bp-1970bp
GTCGATTTATTGGCAGTA
Primer V4 Seq vinc for2 mvinculin, start at 1801bp-1818bp
GATACTACAACTCCTATC
Primer V5 Seq vinc for3 mvinculin, start at 2510bp-2525bp
CTCAGGAGCCTGACTTC
Tabelle 2-1: List of sequencing primers used during this work; bp is an abbreviation for “base pair”.
Materials & Methods
16
Number Name Purpose Sequence 5’ to 3’ Primer 1a pEGFP, for, Nhe I
amplification of pcEGFP-N2 vector
GCGCTAGCGCTACCGGTCG
Primer 1b pEGFP, rev, Xho I
amplification of pcEGFP-N2 vector
GCCTCGAGCTTAAGGAGTCCGGACTTGTACAG
Primer 2a mvinc_N, for, Afl II
amplification of vinculin/ vinculinΔC for pcEGFP-N2
GCCTTAAGATGCCGGTGTTTCACACG
Primer 2b mvinc_N, rev, Xba I
amplification of vinculin/ vinculin-tail858-1066 for pcEGFP-N2
GCTCTAGACTACTGGTACCAGGGAGT
Primer 2c mvinc0.5, for
gene SOEing of vinculin AATAGGGAAGAGGTATTTGAT
Primer 2d mvinc0.5, rev
gene SOEing of vinculin ATCAAATACCTCTTCCCTATT
Primer 2e mvincdelC, rev, Xba I
amplification of vinculin ΔC/ vinculin-tail858ΔC for pcEGFP-N2
GCTCTAGACTAATCTGTTCGGATTTTGATTGA
Primer 2f Vinculin-tail, for, XbaI amplification of vinculin-tail858/ vinculin-tail858 ΔC for pcEGFP-N2
GCCTTAAGATGCTGGCTCCTCCTAAGCCA
Primer 3a Vinculin Y1065F, for Mutagenesis of the C-terminal src phos-phorylation site
CAGAAAGACTCCCTGGGAACAGTAGTCTAGAGGG
Primer 3b Vinculin Y1065F, rev Mutagenesis of the C-terminal src phos-phorylation site
GTCTTTCTGAGGGACCCTTGTCCATCAGATCTCCC
Table 2-2: Primers for cloning and mutagenesis; Note, SOEing is an abbreviation for “splicing by overlap extension”
2.1.4 Vectors
The pcDNA 3.1 Hygro/(+) vector (Invitrogene, Karlsruhe) was used for generating an
enhanced green fluorescent protein (EGFP) vector carrying a hygromycin resistance gene.
The EGFP cassette was amplified via a polymerase chain reaction (PCR) using a pEGFP-
actin vector from BD Bioscience (Heidelberg). The vinculin, vinculinΔC, vinculin-tail,
vinculin-tailΔC and vinculinY1065F were subcloned into the generated pcEGFP-vector. The
vinculin-LD, vinculin-CT and vinculin-H3 were cloned into a pEGFP-C2 (Clontech,
Heidelberg) vector and kindly provided by Dr. Wolfgang Ziegler (University of Leipzig).
2.1.5 Bacterial cultures and cell lines
The cloning was performed in the E. coli strain DH5α with the following genotype: F-,
φ80dlacZΔM15, Δ(lacZYA-argF)U169, deoR, recA1, endA1, hsdR17(rk-, mk+), phoA, supE44,
λ-, thi-1, gyrA96, relA1. The strains were grown in LB-medium, and LB-agar was prepared
according to the manufactor´s describtion. Antibiotics such as Ampicillin (100µg/ml) or
Kanamycin (30µg/ml) were used for liquid medium and plates. The cell lines employed in
this study were mouse embryonic fibroblasts (MEF) isolated from vinculin knock out mice
Materials & Methods
17
(MEF-vin(-/-)).The unaffected control fibroblasts (MEF-wt) were kindly provided by Dr. E.D.
Adamson (La Jolla, CA).
2.1.6 Antibodies
Antibodies were mainly employed for immuno detection in Western Blots and in vivo
imaging of cells by immuno fluorescence. The antibodies used in immunoblots and
immunofluorescence are listed in Table 2.3. The secondary antibodies and toxins used in
these approaches are listed in Table 2.4.
Description Antigen Company Organism Clonal type Anti-vinculin Vinculin-head Sigma Mouse Monoclonal
Anti paxillin Paxillin BD Bioscience Mouse Monoclonal
Anti-Actin Beta-actin Sigma Mouse Monoclonal
Tabelle 2-3: List of primary antibodies used in immuno blots and immuno fluorescence.
List of secondary antibodies used in immuno blots and immuno fluorescene. 2.2 Methods 2.2.1 Methods in molecular biology
2.2.1.1 Quantification of Desoxyribonuclein acid (DNA) DNA concentrations were determined photometrically. The absorption maximum of DNA is
at 260 nm. An absorption value of 1 at the given wavelength of 260 nm is equal to 50 µg/ ml
DNA. The concentration of DNA in a diluted solution is calculated as follows:
OD260
× dilution factor × 50 μg/ ml = Concentration of DNA μg/ ml.
2.2.1.2 DNA electrophoresis The electrophoresis of DNA is an essential technique in molecular biology relying on the
negative charge of DNA for size separation in a sieving matrix. Traditionally, electrophoresis
is performed using agarose (an extract of seaweed) and a Tris(hydroxymethyl)aminomethane
(Tris) buffering solution as described in Sambrook et al [83]. Agarose gels of 1-2 % were
Description Antigen Company Organism Clonal type Alexa FluorTM545 phalloidin
Amanita phalloides Molecular Probes - Monoclonal
HRP-linked Antibody Anti-Mouse IgG Amersham From sheep Monoclonal TRITC Anti-Mouse IgG Jackson
ImmunoResearch Donkey Monoclonal
FITC Anti-Mouse IgG Jackson ImmunoResearch
Donkey Monoclonal
Table 2-4:
Materials & Methods
18
used. The gels were prepared using 1 x Tris-acetate (TAE) buffer pH 8 (40 mM Tris, 20 mM
acetate and 1 mM Ethylendiamintetraacetat ((EDTA)). Between 0.5-1 % of Ethidium
Bromide was added to the gel and the running buffer for visualization of the DNA. DNA
sample buffer (0.25 % Bromphenol blue, 0.25 % Xylencanol, 30 % Glycerin) was added to
the DNA. The DNA was separated at 120 mA. As standard, the 1 kb DNA ladder from New
England Biolabs (NEB, Frankfurt/ Main) was used.
2.2.1.3 Restriction endonuclease digestion of DNA Restriction enzymes were purchased from NEB (Frankfurt/ Main), and the reaction was set up
using the appropriate amount of enzyme in units as recommended by the manufacturer. For
preparative digestion, 2.0-5.0 μg DNA was used. For analytical purposes 0.2 – 0.3 µg DNA
was used. The reaction was performed in the suitable 1 × buffer provided by NEB (Frankfurt/
Main). The reaction mix was incubated for 1-2 hrs at 37 °C, if not otherwise indicated, as
recommended by the manufacturer. The digestion assays were separated and monitored on
agarose gels as described under 2.2.1.2.
2.2.1.4 Ligation The ligation reaction covalently links the phosphordiester bonds between two fragments of
DNA using the ligase enzyme. DNA ligase from T4 Bacteriophage (Boehringer, NEB) was
employed in this work to link vector DNA with the insert DNA. Appropriate molar ratios
(usually 1:3) between vector and insert DNA were used in a 15 μl reaction with 1-2 units
ligase and 1 × T4 ligation buffer. The reaction mixture was incubated at room temperature for
1 hour or overnight at 12 °C, respectively.
2.2.1.5 Preparation of chemical competent DH5α The bacteria were prepared according to a modified protocol of Hannahan [84]. A pre-culture
of DH5α was grown over night in Luria-Bertani (LB) medium. 100 µl of that culture was
transferred into 50 ml LB-medium and inoculated at 37 °C and 220 rpm until an OD600 of 0.4-
0.5 was reached. Prior to centrifugation (2000 g, 7 min at 4 °C), the culture was incubated for
10 minutes on ice. After centrifugation the bacteria pellet was resuspended in 15 ml Calcium/
Magnesium (CM) buffer (50 mM CaCl2, 50 mM MgCl2, 10 % Glycerol), followed by another
incubation step on ice for ten minutes. The bacteria were spun down for 5 minutes at 2000 g
and 4 °C. The pellet was dissolved in 3.5 ml CM and incubated again on ice for 10 minutes.
After the addition of 125 µl Dimethylsulfoxid (DMSO; Sigma, Deisenhofen), the cells were
kept for additional 5 minutes on ice. The bacteria solution was aliquoted and stored at -80 °C
after adding a final volume of 125 µl DMSO followed by 5 minutes of incubation on ice.
Materials & Methods
19
2.2.1.6 Transformation of competent cells The ligated DNA was transformed in chemically competent cells. The E. coli strain DH5α
was used for this purpose. Cells were incubated for 30 minutes on ice with the vector DNA. A
heat shock at 42 °C was applied for 90 s followed by 45 minutes incubation in 0.6 ml LB-
medium at 37 °C and low agitation for cell regeneration. The cells were spun down by
centrifugation at 8,500 g, resuspended in 100 μl medium and plated onto LB agar medium
containing appropriate antibiotics.
2.2.1.7 Polymerase chain reaction (PCR) PCR is a technique employed for rapid in vitro amplification of DNA [83]. The amplification
is done using a temperature resistant DNA polymerase, free nucleotides, a primer or the
starter sequence which specifically binds to the template strand. Three major steps are
involved in the reaction: the denaturation of the template DNA at high temperature (95 °C),
annealing or primer binding to the template strand at 55 °C – 60 °C and finally an extension
step at 72 °C for the actual amplification. For exponential increase of amplified DNA, the
steps are repeated for 25 - 30 cycles. DNA polymerase from thermophilic bacterias such as
Thermus aquaticus (Taq) does not have the 3’ to 5’ exonuclease or “proof reading” properties
to prevent mutations. Therefore, this enzyme was only used for the amplification of small
fragments (<1000 bp) such as the EGFP cassette or the vinculin-tail. The DNA polymerase
from Pyrococcus furiosus (Pfu) was used for amplifying whole vinculin and vinculinΔC
cDNA fragments >1000 bp. The standard PCR reaction mixture contained template plasmid
DNA (50 ng concentration), template specific primers (100 pmol), free nucleotides or dNTP’s
(0.3 mM), MgCl2
(0.5 – 2.5 mM), 1 × PCR buffer and 0.5 - 5.0 units of DNA polymerase
(Taq/Pfu) in 25 – 50 μl reaction volume.
2.2.1.8 Site directed mutagenesis The two step site directed mutagenesis protocol according to Wang and co-workers was
employed to introduce the Y1065F mutation in vinculin [85]. Oligonucleotide sequences
(primer) with the respective bases exchanged were used. In a first step, two separate PCR
reactions, one for the coding and one for the non-coding strand, were performed using only
one mutagenesis primer for each reaction. This first step ensures the efficient introduction of
point mutations. After 5 separate cycles, the PCR assays were combined for additional 25
cycles. The PCR reactions contained the same reagents as described in 2.2.1.7. The PCR
products were digested with the restriction enzyme Dpn I at 37 °C for one hour. Dpn I
specifically digests the methylated parental DNA used as template and retains only the newly
Materials & Methods
20
synthesized DNA. The reaction mixture was then transformed in DH5α competent cells and
plated onto LB agar medium with an appropriate antibiotic selection.
2.2.1.9 Purification of plasmid DNA from bacteria Plasmid DNA amplified in DH5α cells was purified by the modified alkaline lysis method
[83]. Plasmid mini-preparations were performed using the Nucleo Spin plasmid extraction kit
from (Macherey-Nagel, Düren); plasmid maxi-preparations were performed using Qiagen
columns (Qiagen) according to the manufacturer’s recommendation.
2.2.1.10 DNA sequencing DNA sequencing reactions were performed by MWG Biotech according to Sanger [86]. For
that purpose, 1 µg of the plasmid DNA for each reaction together with 10 pmol of the
appropriate sequencing oligonucleotide were sent to the company.
2.2.1.11 DNA extraction from agarose gels DNA was recovered from agarose gel slices with the QIAex Gel Extraction Kit (Qiagen) as
suggested by the manufacturer.
2.2.1.12 Cloning strategy – EGFP expression vector The EGFP expression system was generated as described in figure 2-1. For the generation of
an EGFP expression vector with a hygromycin resistance gene, a pcDNA 3.1 Hygro/(+)
vector (Invitrogene, Karlsruhe) was used. The EGFP cassette was amplified from a pEGFP-
Actin Vector (BD Bioscience) by a PCR reaction and subcloned N-terminally to the multiple
cloning site (MCS) of the pcDNA 3.1 Hygro/(+) vector via Nhe I/ Xho I.
Materials & Methods
21
Figure 2-1: Cloning strategy for the generation of the pEGFP expression system.
2.2.1.13 Cloning strategy – vinculin full length and tail domain constructs The full length mouse vinculin cDNA was kindly provided by Dr. E.D. Adamson (La Jolla,
CA) and Dr. W.H. Ziegler (Leipzig). Sequencing analysis identified several mutations in
these clones. There were three mutations which resulted in amino acid replacements in the
vinculin cDNA provided by Dr. Adamson (F633S, V828E, L936P). The vinculin cDNA
obtained from Dr. Ziegler showed two mutated residues in the vinculin-head region (V246A,
A486V). All those mutations were considered critical for the protein function. The following
strategy was developed to obtain a vinculin cDNA without any mutations. The first part of the
vinculin cDNA from Dr. Adamson was fused with the unaffected part of Dr. Ziegler’s cDNA
using a gene SOEing protocol as described in figure 2-2. In a first PCR reaction, vinculin part
I from Dr. Adamsons cDNA and vinculin part II from Dr. Ziegler’s cDNA were amplified
by overlapping the complement ends. In a second PCR, the products of the first two separate
reactions were combined. They aligned on the complement ends which gave the PCR together
with primer 2a and 2b the possibility to synthesize the strain.
Materials & Methods
22
Figure 2-2: Cloning strategy for the vinculin cDNA. Black arrows mark the positions of the mutated codons. The generated vinculin cDNA was cleaved into the constructed pcEGFP-N2 vector (see
2.3.1.10) using the restriction sites of Afl II and Xba I. The fused vinculin cDNA was used as
template for the generation of vinculinΔC (1-1052), vinculin Y1065F, vinculin-tail (858-
1066) and the vinculin-tailΔC (858-1066) cDNA. The endproduct was also verified by
sequencing.
2.2.2 Cell biology methods
2.2.2.1 Cell culture The mouse embryonic fibroblasts (MEF-wt) and vinculin null fibroblasts (MEF-vin(-/-)) were
grown in Dulbecco´s modified eagle medium (DMEM), 1 g/ l glucose (Biochrom, Berlin)
with 10 % FCS (Biochrom, Berlin) and 1 % Penicillin/ Streptomycin antibiotics at 37 °C and
5 % CO2. For liquid nitrogen storage, 80 µl of DMSO was added to 2 x 105 – 5 x 105 cells
dissolved in 1 ml DMEM with additives.
2.2.2.2 Transient transfection of cells Transfections were carried out with LipofectaimeTM (Invitrogen, Karlsruhe) according to the
manufacturer’s protocol. In brief, 4 μg DNA and 10 µl LipofectaimeTM were dissolved in 250
µl pure DMEM, respectively. After 5 min incubation at RT, the dissolved DNA and
Materials & Methods
23
lipofectaimeTM solutions were combined and incubated for additional 20 min at RT. The
reaction assay was pipetted onto 2 x 105 cells plated in a dish of 3.2 cm diameter
approximately 24 h prior to transfection. The cells were incubated for 4 h at 37 °C with a total
volume of 2.5 ml pure DMEM. After 4 h of incubation, the pure DMEM was replaced against
DMEM with additives as described under 2.2.3.1. The constructs were allowed to express for
16-24 hours prior to further procedures.
2.2.2.3 Culturing of cells on the cover slips for microscopic analysis 1 x 105 – 2 x 105 of transfected or non-transfected cells were seeded onto 15 mm glass
coverslips that had been pre-treated with a mixture of 70 % ethanol. The ethanol on the
coverslips was allowed to evaporate under laminar flow prior to coating. Glass coverslips
were coated with 25 μg/ml fibronectin (Roche, Penzberg) for at least 2 hours at room
temperature or overnight at 4 °C. Uncoated coverslips were used for focal contact
determination.
2.2.2.4 Fixation and permeabilization Cells (transfected and non-transfected) were fixed with 3 % para-formaldehyde (PFA) in PBS
for 15 minutes. After fixation, the cells were permeabilized with 0.1 % Triton X-100 for 5
minutes. The cells were thoroughly washed three times with 1 × PBS and stored at 4 °C until
they were stained.
2.2.2.5 Immunolabelling, fluorescence microscopy and image processing To avoid non-specific interactions, cells were blocked prior to antibody staining with 1 %
bovine serum albumin (BSA, Sigma, Deisenhofen) in phosphate buffered saline (PBS) for 20
minutes. Thereafter, the coverslips were incubated with 50 μl of primary antibody solution on
parafilm for 1 - 2 hours at room temperature followed by 3 washes in 1 × PBS. Staining with
secondary antibody with or without the addition of fluorescently coupled phalloidin was
carried out for 1 hour at room temperature. Coverslips were mounted in 30 μl Mowiol on
microscope glass slides, dried and stored in the dark at 4 °C until analysis. Cells were
observed on a LEICA microscope DMI6000 equipped for fluorescence and phase contrast
using 20 x/0.4NA-, 40 x/0.6NA- as well as 63 x/1.3NA objectives. The immersion oil was
obtained from Leica. Data were acquired by a CCD camera (ORCA ER, Hamamatsu) driven
by the WASABI software package. Data were stored and analyzed using the Image J
software. The cells for focal adhesion determination were examined using a Zeiss LSM 510
META confocal laser scanning microscope system with a 25 x/0.8 NA Plan-Neofluar oil
objective. Images were acquired and processed using Zeiss LSM Image browser (Zeiss, Jena).
Materials & Methods
24
2.2.2.6 Determination of the spreading area Prior to fixation (see chapter 2.2.2.3), transfected and non-transfected MEF cells were
cultured on coverslips for 1h (see chapter 2.2.2.2). The actin cytoskeleton at non-transfected
cells was stained as described under 2.2.2.5. Pictures were acquired at 20x magnification
using a fluorescence microscope. The number of cells and their spreading area computed
using a MatLab image analysis program.
2.2.2.7 Fluorescence recovery after photo-bleaching (FRAP) measurements MEF-vin(-/-) cells were transfected with EGFP-labeled vinculin and vinculinΔC constructs as
described under 2.2.3.2. 1 x 105 of the transfected cells were seeded on fibronectin-coated
(25µg/ml) glass bottom wells. The glass bottom wells with the transfected cells were placed
in the chamber on a temperature-regulated 37 °C microscope stage 30 minutes before
measurements. Time-separated confocal 8-bit images were collected using a Zeiss LSM 510
META confocal laser scanning microscope system with a 63 x Plan-Apochromat oil objective
(NA 1.4). GFP was excited with the 488 nm line of the Argon laser and reflected and emitted
light was separated using a 505 nm long pass filter (for viewing GFP emission). During
bleaching, the 488 nm laser line output was set to 100 % using 50 iterations. Prior to
bleaching, the sample cells were scanned for 60 s to determine the stability of focal contacts.
For determination of the background photobleaching caused by the line scans, at least 3
reference focal contacts and one cytosolic region were selected. The line scans were collected
for at least 5 minutes. To eliminate the background fluorescence at the cytoplasm as well as
incomplete bleaching, the offset cs(t) was calculated as follows:
)((t)c 0s tFad= (1)
where Fad represents the fluorescence intensity of the bleach field at the FA and Fcp the
fluorescence intensity of the bleach field of the cytoplasm. Time point t0 determines the start
of bleaching. The mean of the intensity of the reference focal adhesions were averaged (Fref
(t)) and the bleach function was normalized due to following equation:
0n ))()((
)()((t)F
FtctFtctF
sref
sad
−−
= (2)
The pre-bleach intensity F0 was determined over ten timepoints (tp) before starting the
bleaching according to
spref
spad
ctFctF
−
−=
)()(
F0 (3)
Materials & Methods
25
Prior to analysis, the curvatures were fitted to exponential function
)1(f(t) kte−−= α (4)
Using the MATLAB function fminsearch, whereas α represents the saturation value and k
the turnover rate. The recovery halftime was determined according to
k2lnt1/2 = (5)
2.2.2.8 Electron microscopy measurements The focal contacts in MEF cells were investigated using a transmission electron microscope
(TEM). Dependent on the density of cell compartments, electrons go through the sample and
hit the detector or were scattered on the dense structures which result in a picture. The
resolution of such a microscope is around 1Å. 3 x 105 MEF cells were seeded on a positively
charged “Melinex” plastic strip (2.8 cm x 0.5 cm) deposited in a 3.2 cm cell culture dish. The
plastic strip with the overnight attached cells was transferred into a glass dish prior to fixation.
The cells were incubated for 20 minutes with a fixative consisting of 4 % para-formaldehyde,
0.5 % glutaraldehyde dissolved in 1 M sodium-cacodylate buffer. The cells were washed
twice using 1 x PBS. The sample was sequentially incubated in ethanol solutions of 30 %, 40
%, 50 %, 60 % 70 %, 80 % 90 % and 100 %. Each step took 10 min and occurred at RT.
Afterwards the cells were treated with the London resin white (LRW; Sigma, Deisenhofen) by
incubating them sequentially for 30 minutes at room temperature with a mixture of
ethanol/LRW (1 : 1), ethanol/LRW (1 : 3) followed by 2 steps of pure LRW incubation. The
strip with the cells was embedded into a gelatine capsule using LRW. The resin was hardened
by incubating the capsule for 24 h at 60 °C. The gelatine shell was removed by a razor blade,
the LRW block splitted longitudinal to the cells and the plastic strip was ripped of as
illustrated in figure 2.3. The resin block with the cells was re-embedded in a second gelatine
capsule and filled with additional LRW resin to prevent the loss of the cells during sectioning.
After additional incubation for 24 h at 60 °C ultra-thin sections of 70-100 nm thickness were
prepared using a microtom. The sections were deposited on a copper grid. Before immuno-
labelling, the sections were blocked by a 1 % BSA solution dissolved in filtered 0.01 M PBS
(pH 7.4). Thereafter, the samples were incubated with the primary antibody solution for 1 h at
room temperature followed by 4 washes in 1 × PBS. Staining with gold-labeled secondary
antibody was carried out for 30 minutes at room temperature. After washing the samples with
filtered 0.01 M PBS, the sections were jet-washed for 30 seconds with distilled water. Finally,
the sections were incubated for 7 minutes with 4 % uranyl-acetate followed by 4 minutes of
Materials & Methods
26
incubation with lead to improve the contrast during the measurement. The samples were
investigated using a JEOL (JEM – 2000FX) TEM.
Figure 2-3: Sample preparation for TEM measuremnts. (a) Melinex plasic strip bottom, (b) Melinex plastic strip top. After the LRW hardening, the crystal was separate longitudinal to the plastic strip and the plastic strip was ripped off. The crystal part with the cells (A) was re-embedded into another gelatine capsule and after hardening used for preparation of the microsections.
2.2.3 Biochemical and biophysical methods
2.2.3.1 Bead coating
Superparamagnetic 4.5 µm epoxylated beads (Invitrogen, Karlsruhe) were coated with
100µg/ml fibronectin (Roche, Mannheim) in PBS at 4 °C for 24 h. Beads were washed in
PBS and stored at 4 °C. The Integrin receptors bind to the RGD motif, localized in FN.
2.2.3.2 Peptide synthesis
The different vinculin peptides, representing vinculin´s C-terminus were kindly provided by
Dr. Rothemund and Dr. W.H Ziegler (University of Leipzig). The peptides, listed in Table 2.4
were generated according to the method of Marryfield [87]. Peptides were synthesized by
coupling the activated backbone carboxyl group of one amino acid to the backbone amino
group of another residue. The possibility of unintended reactions is obvious, thus protecting
groups are usually necessary.
Vinculin Peptide (residue 1045-1066) Sequence
C-terminal arm (residue 1045–1066) IKIRTDAGFTLRWVRKTPWYQ
C-terminal arm R1060, K1061 to Q IKIRTDAGFTLRWVQQTPWYQ Table 2-5: Vinculin´s C-terminal peptides used in this work
2.2.3.3 Lipid vesicle preparation Multilamellar vesicles (MLVs) were prepared as follows: lipid stock solution consisting of
Dimyristoyl-L-α-phosphatidylcholin (DMPC) and Dimyristoyl-L-α-phosphatidylglycerol
Materials & Methods
27
(DMPG) were purchased from Avanti Polar Lipids (Birmingham, AL). Mixtures of
crystalline DMPG and DMPC were dissolved in chlorophorm/methanol, 2 : 1 (v/v). To form a
dry lipid film on the glass wall of an Erlenmeyer-flask, the solvent was evaporated by
nitrogen followed by a vacuum desiccation step for at least 2 h. The lipid film was suspended
in a lipid buffer containing 20 mM HEPES (pH 7.4), 2 mM EDTA, 5 mM NaCl and 0.2 mM
DTT for generating multilamellar vesicles. The vesicles equilibrated overnight at 35 °C.
For the CD-spectroscopy measurements small unilamellar vesicles (SUVs) were prepared.
For this purpose, the lipid film in the Erlenmeyer-flask was dissolved in 10 mM potassium
phosphate buffer (pH 7.4) and sonified for 10 minutes after the equilibration step.
2.2.3.4 Differential scanning calorimetry (DSC) measurements DSC is a technique to resolve conformational transitions of biological macromolecules by
energy determination. It provides an immediate access to the thermodynamic mechanism that
governs a conformational equilibrium, i.e. between folded and unfolded forms of a protein or
DNA. The theory of DSC and the thermodynamic interpretation of the experiment have been
the subject of excellent reviews {Goldmann, 2008 #953]. In brief, the DSC registers phase
transition energy of biological macromolecules and can be used to determine the heat capacity
calorimetrically. The enthalpy of an isobaric system is defined as
ΔQΔH = (6)
where, H is the enthalpy and, Q the heat energy. Furthermore, under these conditions, the heat
capacity can be described as
ΔTΔHCp = (7).
Integration of the determined heat capacity (Cp) over the change in temperature gives the
phase transitions enthalpy (H):
∫=fluid
gelpdTCΔH (8).
The DSC experiments in this work focused on the phase transition enthalpy of phospholipids
in presence and absence of peptides. The principle is described in figure 2-4. Phospholipids in
membranes can exist in a gel-like (ordered) as well as fluid-like (disordered) phase dependent
on the temperature. At the ordered phase, all acyl-residues of a fatty acid are in the
energetically favourable trans-conformation. Transferring the phospholipids into the fluid-like
Materials & Methods
28
phase by heating up the system constantly, leads to a change in conformation and allows the
energetically unfavourable gauche conformation of the acyl residues. This results in a
disturbed package of the acyl-residues in the vesicles which is responsible for the transition to
the liquid state. The phase change is regarded as first order phase transition. The energy,
necessary for that conformational change could be determined by the specific heat capacity
(Cp), also called specific heat.
Figure 2-4: Phase transition thermogram of DMPG/DMPC vesicles. Both pre-transition and transition occur with increased temperature. TV marks the maximum temperature of the pre-transition and, TM represents the melting point for the main phase transition, where the specific heat has the calorimetric maximum. TS and TL mark the start and end points of the transition. As the high energy (gauche) conformations increase, the energetically favourable (trans) conformations decrease during phase transition [88]. Some lipids show a pre-transition phase at which the gel-like membrane changes from a
lamellar (Lβ´) to a ripple phase (Pβ´), before undergoing the transition to the fluid phase (Lα) at
the melting temperature (TM). A polypeptide with the ability to insert into a lipid membrane
interacts non-covalently with some of the acyl-residues and, therefore, prevents phase
transition for these lipids. This results in a reduced TM and a decreased phase transition
enthalpy (H).
The heating of the sample and reference solution is performed at a preset heating rate, ß = ΔT/
Δt. The principle of the DSC requires the temperature of the sample (TP) and the reference
(TR) solution to remain constant (T = TP = TR) as described in figure 2-4. To ensure an
endothermic phase transition, the sample solution must be heated to a higher temperature than
Materials & Methods
29
the reference solution to keep both temperatures constant. The heat output for the sample (Pp)
will be higher than for the reference (PR). This is reflected in the heat capacity ΔC (T)
between sample and reference ΔC (T) = Cp - CR =ΔP (T) / ß [88].
Figure 2-5 A schematic of the DSC apparatus with the sample (1), reference (2), heater (3), insulation coat (4) and temperature observation points (5) and PP and PR the probes of the sample (P) and the reference (R) respectively [88].
A differential scanning calorimeter Q100 from TA Instruments was used for all
measurements. The peptide of vinculin’s lipid binding region and insulin (purchased from
Sigma, Deisenhofen) were dissolved in a lipid buffer (see 2.2.3.3). The lipid buffer solution
was placed in the reference cell and lipid peptide solution in the sample cell. MLVs were
used. Under sealed conditions, both solutions were heated at a rate of 0.5 °C/ min and cooled
at a rate of 1 °C/ min. The heat capacity was captured between +7 °C and +30 °C until the
equilibrium of the phase transition enthalpy was reached. A phase transition was observed at
~23 °C depending on the molar ratio of DMPG and DMPC. Data analysis was performed
using the software from Universal Analysis 2000 (TA Instruments) and Origin 7G.
2.2.3.5 Circular dichroism (CD) spectroscopy measurements Circular dichroism (CD) is a spectroscopic technique which measures differences in the
absorption of left-handed and right-handed circularly polarized light of optical active
molecules such as chiral compounds. Amino acids and so peptides are photoactive molecules.
The Cα atoms of amino acids are achiral. This leads to a disproportionated distribution of the
peptide bond electrons, which is responsible for the difference in absorption of right and left
polarized light. The conformation or secondary structure of the peptide influences this
electron distribution. This results in certain spectra for typical secondary structures as alpha
Materials & Methods
30
helices and beta sheets. A CD spectrum is measured typically between 180 – 300 nm. The
first part is reflecting the absorption of the peptide bond (180 – 230 nm). At 240 – 300 nm,
the aromatic residues give the signal. The difference in absorbtion of left and right polarized
light is blotted as ellipticity (θ).
The 21 residue peptide of vinculin was dissolved in a 10 mM potassium phosphate buffer (pH
7.5). The concentration of the peptide was determined photometrically using an extinction
coefficient of 12,490 M-1 cm-1 at 280 nm. CD spectra were recorded on a JASCO J-815 CD
Spectrometer at 1 nm intervals over the 180 - 260 nm wavelength range using a 70 µM
protein solution. The spectra of the lipid/protein mixture were recorded at a lipid/protein
molar ratio of 40 : 1. Peptides and lipid vesicles were dissolved in 10 mM potassium
phosphate buffer (pH 7.5) and measured in a cuvette of 0.1 cm path length. Three scans were
recorded for each sample and averaged. The protein spectra were corrected by subtraction of
the spectrum of 10 mM potassium phosphate buffer (pH 7.5) or the spectrum of lipid vesicles,
respectively. The corrected protein spectra were adjusted for protein concentration and path
length to obtain the mean residue molar ellipticity at each wavelength.
CD spectra of the vinculin peptide 1045-1066 were obtained at 30 °C and smoothed with the
Savitzky – Golay algorithm. Secondary structure analysis was performed using the
CONTINLL, CDSSTR algorithms provided by DICHROWEB [89-90]. The quality of the fit
between experimental and back-calculated spectrum corresponding to the derived secondary
structure element fractions was assessed from the normalized root mean square deviation
(NRMSD).
2.2.3.6 Solid state NMR Among the biophysical techniques that allow the investigation of peptides and proteins in
bilayer environments solid-state NMR spectroscopy has proven to be a valuable tool. The
technique resolves the structure, dynamics and topology of membrane-associated
polypeptides [88]. The tilt angles of helices with respect to lipid bilayers have been
determined using static-oriented samples. It has been demonstrated that this approach has also
been shown to be suitable for the complete structure determination of membrane-bound
peptides by measuring a large number of conformational constraints. If the lipid peptide
interactions are oriented, NMR interactions are used to extract angular constraints from such
static-oriented samples. Proton-decoupled 15N solid-state NMR spectroscopy of peptides
labeled at the backbone amides with 15N has been proven particularly convenient as this
method provides the approximate tilt angle of membrane-associated helices in a direct
manner. Whereas transmembrane helical peptides exhibit 15N chemical shifts around 200
Materials & Methods
31
ppm, sequences oriented parallel to the surface resonate at frequencies < 100 ppm (Figure 2-
6). Due to fast axial rotation of the phospholipids around their long axis the 31P chemical shift
is characterized by an averaged symmetric tensor. The singular axis (σ||) coincides with the
rotational axis, i.e., the bilayer normal. In the 31P solidstate NMR spectra of pure liquid
crystalline phosphatidylcholine bilayers the signal at 30 ppm is thus indicative of
phosphatidylcholine molecules with their long axis oriented parallel to the magnetic field
direction, whereas a –15 ppm 31P chemical shift is obtained for perpendicular alignments. In
perfectly aligned samples the phospholipid bilayer spectra consists of a single line. Routinely
the 31P NMR spectra of phospholipid bilayers of the peptide carrying samples are recorded to
test for the quality of order and alignment of phospholipid bilayers.
Figure 2-6: Simulated solid-state NMR spectra. (A) and (B) show simulated 15N solid-state NMR spectra of an a-helical polypeptide oriented with the helix long axis perpendicular (A) or parallel (B) relative to the bilayer surface. The membranes are aligned with their normal parallel to the magnetic field of the NMR spectrometer (B0) [88].
NMR measurements were performed under supervision of Dr. Philipe Bertani at the
University of Strassbourg. The vinculin peptide representing the last 21 residues with the
sequence IKIRTDAGFTLRWVRKTPWYQ was used. At the underlined positions the 15N-
labeled analogue of leucine (L) was used (again). Typically, 10 mg of the polypeptide was
resolved in 50µl formic acid. A phospholipids solution consisting of 30 mg 1-Palmitoyl-2-
Oleoyl-sn-Glycero-3-Phosphoglycerol (POPG) and 70 mg 1-Palmitoyl-2-Oleoyl-sn-Glycero-
3-Phosphocholine (POPC) was dissolved in Hexafluor-2-Isopropanol. Both solutions were
mixed and dried onto approximately 30 ultra-thin cover slips (9-22 mm) using a dessicator.
After the complete evaporation of the organic solvents the samples equilibrated at 93 %
relative humidity. The glass plates are then stacked on top of each other, which results in
small brick-shaped samples of 3–4 mm thickness. These are stabilized and sealed with teflon
tape and plastic wrappings. The sample was placed into a coil which was placed that the
Materials & Methods
32
membrane was aligned parallel to the magnetic field direction. Cross-polarization or Hahn
echo NMR pulse sequences are typically used to acquire 15N and 31P NMR spectra. The
measurements were performed in a 400MHz Burker Avance solid state NMR spectrometer
using a magnetic field of 9.4 T.
2.2.3.7 Cell lysis The cells were seeded onto fibronectin (FN) coated cell culture treated dishes of 3.2 cm
diameter 24 h prior to lysis. The attached cells were put in 100 µl radio immuno precipitation
assay (RIPA) –buffer (50 mM Tris-Cl pH 7.4, 150 mM NaCl, 1 % NP40, 0.25 %
sodiumdeoxycholate, 1 mM PMSF) containing 1 mM sodiumvanadate and 1 x CompleteTM
protease inhibitor (Roche, Basel, Switzerland). After washing with 1 x PBS, the RIPA treated
cells were removed from the surface using a cell scratcher. Lysis was enhanced by agitating
the dissolved cells for 15 minutes and 4 °C. For separating the cell debris from the cytosolic
components, the lysate was centrifuged at 4 °C and 14,000 g for 15 minutes. The supernatant
was diluted with 6 x SDS sample buffer and boiled at 95 °C for 3 minutes to prevent further
protein degradation.
2.2.3.8 Protein concentration determination according to Bradford The protein determination according to Bradford is based on the interaction of the Coomassie
dye with the peptides [91]. The built peptide-Coomassie complex causes a shift of the dye´s
absorbtion pattern from 465 nm to 595 nm under acidic pH conditions because of the
interaction between the sulfonate groups of the Coomassie dye and the cationic or apolar
peptide residues. The sensitivity of the assay is between 50 - 500 ng/ ml of peptide. The
Bradford assay was performed according to the manufacturer’s instructions. The peptide
concentration was determined using the absorbtion values at 595 nm of a reference
measurement with BSA protein of different known concentrations.
2.2.3.9 Western blot analysis Sodium-dodecylsulphate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed
with a protocol adapted by Laemmli [92]. Separating gels of 10 % were mostly employed in
this work. The protein solutions or cell lysates were diluted in 6× SDS sample buffer, boiled
at 95 °C for 5 minutes and loaded into gels. “High molecular weight marker” (HMW) from
Paclab was used to determine the molecular weight. Proteins were transferred from the gels
onto a nitrocellulose membrane using a semi-dry blotting apparatus from Biometra. The
membrane, gel and Whatman filter papers were equilibrated in the glycine-methanol buffer
(25 mM Tris pH 8.5, 150 mM glycine, 10 % methanol). The proteins were transferred from
Materials & Methods
33
the gel (negative electrode) to the membrane (positive electrode) at 25 V for 40 minutes. The
efficiency of the transfer was monitored by staining the membranes with PonceauS solution
(0.5 % PonceauS, 40 % methanol, 15 % acetic acid; Sigma, Deisenhofen).
Membranes were blocked for 30 minutes to 1 hour at room temperature or at 4 °C overnight
in blocking buffer (5 % dry-milk in TBS-T buffer (20 mM Tris-HCl pH 7.6, 140 mM NaCl,
0.1 % Tween 20)). Afterwards membranes were incubated for 1 h at room temperature or at 4
°C overnight in TBS-T buffer containing the primary antibody. Membranes were washed
three times for 15 minutes in TBS-T and incubated for 1 hour at room temperature in TBS-T
buffer containing the secondary antibody followed by another 4 washes as described above.
Membranes were then incubated for up to 5 minutes with the chemiluminescence substrates
provided by Amersham Biosciences and exposed to Hyperfilm ECL (Amersham Biosciences)
for 10 seconds – 5 min. Films were developed manually using the Kodak developer and fixing
solutions.
2.2.4 Computational methods
2.2.4.1 Molecular dynamics (MD) simulations One of the principle tools in the theoretical study of biological macromolecules is the method
of molecular dynamics simulations. This computational method calculates the fluctuation and
conformational as well as the energetically changes of peptides and nucleic acids in a time-
dependent manner. The theory has been described in detail [93]. In brief, the MD simulation
calculates the accelerated motion of all atoms in a system during a certain time period
according to a defined force field. This force fields represents the energy values of the
covalent bond as the binding-angles and –distances and the non-covalent bond as the van-der-
Waals and the Coulomb interactions. The resulting forces are then applied to each atom
localized in the system using Newton´s law of motion
ma=F (9)
where (F) is the force and (a) the acceleration which can be described as
dtdv
=a (10)
The integration of equation (9) over the time (t) results in the velocity (ν), described in
0)( vattv += (11)
Materials & Methods
34
Further, ν can be expressed as
dtdx
=v (12)
where (x) represents the position. Integrating equation (10) leads to
0)( xvttx += (13)
that give the position of a particle after a certain timepoint. Due to time scales of molecular
movements the integration time-steps are in the range of 1 - 2 femto seconds (fs). Starting
structures on the atomic level are necessary for the MD simulations. Crystal structures of
peptides determined via X-ray and NMR deliver the appropriate coordinates for the start of a
simulation. As starting structure for all molecular dynamic runs for secondary structure
determination of vinculin´s C terminus, the last 20 amino acids were extracted from the
crystal structure of chicken vinculin (PDB 1ST6). For adding the last absent amino acid (Q),
swissprot PDB viewer (http://www.expasy.org/spdbv/) was used. The entire sequence was
IKIRTDAGFTLRWVRKTPWYQ. All simulations were carried out with the GROMACS
3.3.1 package [94-95]. The simulations with the peptide representing vinculin´s last 21
residues were run on a 4 Processor HPC unit with 2.2 GHz for each Processor and 6.2 GB
RAM using Linux. Simulations of the vinculin-tail variants (residues 890-1066 or residue 890
- 1052) were also pre-processed on the 4 Processor HPC system. The final runs were
performed with 16 Xeon 5160 "Woodcrest" CPUs (3 GHz) on a cluster consisting of 217 nods
using Linux. All simulation were performed in an octahedron box filled with approximately
7850 spc216 water molecules [96] by using periodic boundary conditions. The molecule was
centred in the box and the distance to each border was 1.5nm. For keeping the net charge
neutral during the simulations, we added CL- ions to the octahedron box. The GROMACS
Force field ffG53a6 was used [97]. All bonds were constrained using LINCS [98]. Long range
electrostatics were handled using the Particle Mesh Ewald (PME) method. A non bonded cut-
off of 0.9 nm for Lennard - Jones potential was used. The runs were performed at 300 K by
using a pressure coupled Berendsen temperature bath [99]. Energy minimization with a
tolerance of 1000 kJ mol-1 nm-1 was carried out, using the steepest descent method. Position
restraint was used to equilibrate the system for 50 ps. The final MD runs were carried out with
time steps of 2 fs. Snapshots were taken every 100 ps.
2.2.4.2 Analysis of molecular dynamics simulations The dictionary of secondary structure prediction (DSSP) was used for secondary structure
calculation [100]. The program performs its sheet and helix assignments solely on the basis of
Materials & Methods
35
backbone-backbone hydrogen bonds from the generated trajectories. The DSSP algorithm
defines a hydrogen bond when the bond energy is below -0.5 kcal/ mol from a Coulomb
approximation of the hydrogen bond energy. The structure assignments are defined in a way
that visually appealing and unbroken structures result. In case of overlaps, alpha-helix is
given first priority, followed by beta-sheet.
2.2.4.3 Cluster analysis The snapshots 1 - 100 of the vinculin peptide representing vinculin´s last 21 residues were
used for a cluster analysis. A pair-wise difference Euclidian distance matrix was generated
using the available sixty dihedral angles (20 peptide bonds x phi, omega and psi). The
difference between dihedrals (Δθn,m) was defined as
Δθn,m = min ( | θn,p - θm,p |, | θn,p - θm,p + 360° |, | θm,p - θn,p - 360° | ) (14)
for any two structures n and m for a given dihedral angle p, where p = 1 to 60. Ward’s
geometric (cluster centered), minimal variance, agglomerative hierarchical clustering was
applied to each pair-wise matrix [101]. The resulting dendogram was partitioned using
Mojena’s Stopping Rule (No.1) to identify significantly different clusters, representing
distinct conformational groups [102].
2.2.4.4 Visualization of calculated structures All the results were visualized using swissprot (http://www.expasy.org/spdbv/) and the NOC
PDB viewer (http://en.bio-soft.net/3d/noc.html).
2.2.5 Cell mechanical methods
2.2.5.1 Magnetic tweezer measurements The principle of the magnetic tweezers device used here was described in Kollmannsberger et
al. [103]. Prior to measurements, fibronectin-coated beads (see 2.2.3.1) were sonicated, added
to cells (1 x 105 beads/dish), and incubated for 30 min at 5 % CO2 and 37 °C. A magnetic
field with a high field gradient was generated using a solenoid (250 turns of ∅ 0.4 mm copper
wire, solenoid length = 2 cm, mean solenoid diameter 1 cm) with a needle-shaped core
(HyMu80 alloy, Carpenter, Reading, PA). The needle tip was placed at a distance of 20 - 30
µm from a bead bound to a cell using a motorized micromanipulator (Injectman NI-2,
Eppendorf, Hamburg, Germany). Bright-field images of the cell, bead, and needle tip were
Materials & Methods
36
taken by a CCD camera (ORCA ER, Hamamatsu) at a rate of 40 frames/ s. The bead position
was tracked on-line using an intensity-weighted center-of-mass algorithm. A preset force was
maintained by continuously updating the solenoid current or by moving the solenoid so that
the needle-tip to bead distance was kept constant. Measurements on multiple beads per well
were performed at 37 °C for 1 h, using a heated microscope stage on an inverted microscope
at 40x magnification (NA 0.6) under bright-field illumination. Transfected MEF (-/-) cells had
been identified in the fluorescence mode using an EGFP-filter.
To ensure that cells had not experienced any significant forces resulting from previous
measurements, the needle was moved by at least 0.5 mm between two measurements. The
bead position, the needle tip position and the solenoid current was continuously recorded at a
rate of 40 s-1. Image acquisition was triggered and synchronized with the solenoid current
generator using a custom-made C ++ program run on a PC equipped with an AD-DA board
(NI-6052E, National Instruments).
When a force step with amplitude ΔF was applied to a bead, it moved with a displacement
d(t) towards the needle tip. The ratio d(t)/ΔF defines a creep response J(t) that for all force
amplitudes was well described by a power-law
J(t) = a(t/t0)b (15)
where t0 is a reference time that was arbitrarily set to 1 s. The parameter a (units of µm/nN)
characterizes the elastic cell properties and corresponds to the compliance (i.e. inverse of
stiffness). The exponent b reflects the viscoelasticity of the force-bearing structures of the cell
that are connected to the bead. The numbers of b are between 0 and 1, whereas 0 reflects an
elastic and 1 a viscos behavior [65; 103]. The parameter a changed with the amplitude of the
applied force, indicating a force-dependent non-linearity of the creep modulus: a decreased
with increasing force. The parameter a of the creep response were replaced by an arbitrary
force-dependent functions a(F) and to describe the force-dependence of the creep response,
J(t,F) = a(F) (t/t0)b(F) (16)
The force-dependent differential elastic modulus, 1/a(F), can be easily evaluated at discrete
forces if the force protocol follows a staircase-like pattern. A force protocol with nearly
logarithmically spaced force steps according to 0.5, 1, 1.5, 2, 3, 4, 5, 6, 8, and 10nN was
found to be most versatile, each step lasting for 1 s. In this way, a value for a(F) was obtained
Materials & Methods
37
for every force level. The median correlation coefficient between the fit and the data was r2 =
0.99.
2.2.5.2 2D-traction microscopy measurements Gels (6.1 % acrylamide/ 0.24 % bis-acrylamide) for traction experiments were prepared on
rectangular 75 x 25 mm nonelectrostatic silane-coated glass slides according to the procedure
described by Pelham and Wang [33]. The Young’s modulus of the gels was measured with a
magnetically driven plate rheometer and found to be 12.8 kPa. Red fluorescent 1 µm
carboxylated beads (Molecular Probes) were suspended in the gels and centrifuged at 300 g
toward the gel surface during polymerization at 4 °C. These beads served as markers for gel
deformations. The surface of the gel was activated with Sulfo-SANPAH (Pierce
Biotechnology, Rockford, IL) and coated with 50 mg/ml bovine collagen G (Biochrom). The
cell suspension added on the gel was contained within a silicone ring (flexi-perm, in Vitro,
Göttingen, Germany) attached to the glass slide. Cell tractions were computed from an
unconstrained deconvolution of the gel surface displacement field measured before and after
cell detachment with 80 µM CytochalasinD and Trypsin/EDTA (0.25/ 0.02 %) in PBS [104].
During the measurements, the cells were maintained at 37 °C in humidified atmosphere. Gel
deformations were estimated using a Fourier-based difference-with-interpolation image
analysis [105].
Results
38
3 Results I. In vitro 3.1 Differential scanning calorimetry (DSC) measurements The membrane interaction of vinculin influences cell adhesion and migration [81-82]. There
are three regions localized at the 30 kDa tail domain (residues 858-1066) that have been
identified as candidates for lipid membrane interaction [38]. Two of them are well
characterized alpha-helical regions. The third one is localized at the C-terminal arm which
remained unstructured during crystallization [78]. Pull-down experiments with artificial lipid
vesicles incubated with a vinculin-tail variant lacking the last 15 residues (vtailΔC), so called
lipid anchor, demonstrated that this variant was not able to interact with acidic lipid
membranes [78; 81]. However, it was not determined to what extent the lipid anchor (residues
1052-1066) is involved in lipid interaction.
In this experiment the lipid-binding ability of the C-terminal arm, which includes the lipid
anchor of vinculin, was explored and characterized using differential scanning calorimetry
(DSC). Multilamellar lipid vesicles (MLVs) consisting of DMPC/ DMPG at various molar
ratios were incubated with peptides representing different variants of vinculin´s C-terminal
arm (refer to 2.2.3.4). Since peptides have the ability to insert into a lipid membrane and
interact with the hydrophobic acyl chains, the specific heat capacity and the phase transition
enthalpy (H) will decrease. Insulin which has approx. the same molecular weight (5,8kDa) as
the synthesized vinculin peptides was used as negative control. It is known that insulin is not
inserting into lipid membranes.
3.1.1 DSC measurements of the C-terminal arm peptide
As demonstrated in figure 3-1, the specific heat was determined from MLVs at a molar ratio
of 70:30 DMPC:DMPG that were incubated with the 21 residue vinculin peptide, representing
the C-terminal arm. With increasing peptide concentration (from 0 to 180 µM), the specific
heat of MLVs decreased compared to MLVs in the absence of vinculin peptide (figure 3-1A).
The relative flattening of the curve together with the shift of Tm to lower temperatures are
indicative for peptide insertion into lipid bilayers [106-107]. As a control, DSC measurements
in the presence of lipid vesicles and insulin at different molar ratios were performed. With
increased insulin concentration, the specific heat did not alter significantly (figure 3-1B). The
decrease in relative transition enthalpy (dH/ dH0) as a function of increasing peptide
concentration was plotted for DMPC/ DMPG vesicles of different molar ratios (figure 3-1C).
Results
39
Increasing the concentration of negatively charged lipids in vesicles caused a decrease in
peptide-lipid insertion. Insulin (control) showed no changes in relative transition enthalpy.
These results suggest that the C-terminal peptide of vinculin is directly involved in membrane
interaction of vinculin.
Figure 3-1: The peptide, representing vincuin´s C-terminal arm (residue 1045-1066), interacts with MLVs. (A) Thermograms from DSC measurements with lipid vesicles containing DMPC and DMPG at a molar ratio of 70:30. The specific heat decreased with increasing peptide concentration or the peak expanded (n = 4–6 runs). Note, that the integral of the specific heat provides the phase transition enthalpy dH. (B) The phase transition enthalpy of insulin incubated with DMPC/DMPG vesicles of a molar ratio of 70:30. The specific heat remained almost constant with increasing the insulin concentration. (C) Relative phase transition enthalpy changes of DMPC/DMPG vesicles at different molar ratios. With increasing vinculin peptide concentration, the phase transition enthalpy decreased as a consequence of the peptide insertion into the lipid membrane. Note, that there were no detectable changes in enthalpy at all lipid compositions for insulin (data not shown).
3.1.2 DSC measurements of the mutated C-terminal arm
Pull-down experiments with artificial lipid vesicles revealed that the mutation of the basic
amino acids Arginine 1060 and Lysine 1061 to Glutamine (vtail-RK 1060/61 Q) results also
in impaired lipid vesicle interaction [82]. To test the nature of this lipid interaction, the
peptide representing this altered C-terminal arm (RK 1060/61 Q) was also measured using
DSC. With increasing peptide concentration, the specific heat of MLVs incubated with RK
1060/61 Q also decreased in comparison to the measurements performed with MLVs only,
Results
40
whilst Tm was shifted to lower temperatures (figure 3-2A). The relative transition enthalpy
(dH/dH0) for DMPC/DMPG vesicles at different molar ratios (figure 3-2B) also decreased;
even more, compared to the measurements of the non-mutated peptide (for comparison see
also figure 3-1C). Decreasing the concentration of negatively charged lipids, the insertion of
the peptide was also increased, judged by the relative transition enthalpy (see figure 3-2B).
The results suggest that the RK 1060/61 Q peptide of vinculin showed a better membrane
insertion behavior than the wildtype peptide.
Figure 3-2: Interaction of the mutated vinculin-tail peptide (RK1060/61 Q) with MLVs. (A) Thermograms from DSC measurements with lipid vesicles containing DMPC and DMPG at a molar ratio of 70:30. The specific heat decreased or expanded with increasing peptide concentration (B) Relative phase transition enthalpy changes of DMPC/ DMPG vesicles at different molar ratios of DMPC/ DMPG. The increasing vinculin peptide concentration results in a decreased phase transition enthalpy as a consequence of the peptide insertion into the lipid membrane.
3.2 Molecular dynamics simulations Proteins and peptides are highly dynamic structures. Their function is determined by their
secondary structure. The interaction with substrates or other proteins induces conformational
shifts which are important for their functions. Crystal structure analysis delivers only a static
picture of such molecules. In contrast, molecular dynamic investigations give an indication
for proteins in motion during their function.
3.2.1 MD Simulations of the C-terminal arm
Crystal structure analyses of the vinculin molecule demonstrated that the C-terminal arm
remained unstructured and can be divided into three different regions: a flexible loop
(residues 1047–1052), a beta-clamp (residues 1053–1061) and a hydrophobic hairpin
(residues 1062–1066) [78]. As demonstrated in pull-down experiments, parts of this C-
terminal arm known as the lipid anchor (residues 1052–1066) influence the membrane
binding of vinculin [78; 81]. In order to elucidate the secondary structure of the mostly
Results
41
unstructured C-terminal lipid binding region of the vinculin molecule, molecular dynamics
simulations were carried out using the GROMACS package (see also chapter 2.2.4.1). The
MD calculations were performed under different ionic conditions representing basic, neutral
and acidic pHs. The trajectories of vinculin´s last 20 residues were extracted from the crystal
structure of chicken vinculin (PDB 1ST6) [55]. Because of the high flexibility of the C-
terminal arm during crystallization, the position of vinculin´s last residue is missing. So, the
last glutamine (Q) residue was added. The N-terminal isoleucine was always considered as –
NH2 (uncharged) and the five basic residues, the single acidic residue and the terminal
carboxyl-group were either protonated or deprotonated, respectively, depending on the
nominal pH conditions. Under “neutral” conditions, the five basic residues were protonated
and both, the single acid residue and the terminal carboxyl group, deprotonated. Under the
“acid” condition both carboxyl groups were protonated. During the “basic” condition the five
basic residues were completely deprotonated. The nominal charges were therefore -1, 0 or +1.
Charge neutrality of the entire system was achieved by adding either Cl- or Na+ counter ions.
Three 10 ns runs of each condition were performed. The calculated conformations were
collected every 100 ps. The program dictionary of secondary structure prediction (DSSP)
[100] was used for identifying the secondary structure elements of each run (refer to 2.2.4.2).
Further, a conformational analysis of all runs was performed under nominal “basic”, “neutral”
and “acidic” conditions to identify a representative secondary structure. Each simulation was
repeated three times to proof the reliability of the secondary structure determination. The
results are plotted in figure 3-3, 3-4 and 3-5. Under nominal “basic” (Figure 3-3A) and
“neutral” (Figure 3-4A) conditions, an anti-parallel beta-sheet emerged after 3.5 ns and
remained constant until the end of the simulation at 10 ns. Residues 2-12 were part of that
secondary structure element (illustrated in red). Residues 13-21 remained unstructured under
those conditions. Under basic conditions, all three runs delivered the same result whilst only
two of three runs delivered the anti-parallel beta-sheet under “neutral” conditions. The
simulation, performed under nominal “acidic” conditions, obtained no defined secondary
structure elements (Figure 3-5A). Ward’s geometric (cluster centered), minimal variance,
agglomerative hierarchical clustering was applied to perform the cluster analysis for the
snapshots 1 to 100 under all nominal pH conditions. The resulting dendograms were
partitioned using Mojena’s Stopping Rule (No.1) to identify significantly different clusters
[102]. The analysis as demonstrated in figure 3-3B revealed 6 statistically significant
conformational groups (p<0.05) under nominal “basic” conditions. The clusters on the
diagram are illustrated in red and black, respectively. The representative geometry for each
Results
42
cluster is colored in green. The cluster labeled with an asterisk is the most densely packed one
and represents the most energetically favorable cluster. Where the snapshots appear to jump
between clusters, the associated conformations are likely to lie on a plateau, and not a well-
defined local minimum, on the potential energy landscape for the simulation. The cluster
analysis for the run under “neutral” conditions delivered 4 distinct groups (Figure 3-4B)
whilst the simulations under nominal “acidic” (Figure 3-5B) conditions obtained also 6
distinct groups. The structures extracted from the most densely packed groups represent the
most energetically favorable structure of the run and can be seen as the representative
structure of the simulation. Under nominal “basic” (figure 3-3C) and “neutral” (figure 3-4C)
conditions, the extracted structure showed an anti-parallel beta-sheet followed by a mostly
unstructured part. The “acidic” run revealed an undefined secondary structure (figure 3-5C).
These results suggest that the first 6 residues of the lipid anchor (residues 1052-1057) are part
of the turn and the second strand of the beta-sheet. This implies an involvement of vinculin´s
third lipid binding site (residues 1052-1066) in beta-sheet formation. Furthermore, the
simulations indicate a pH-dependent bi-stable behavior. The presence and absence of the beta-
sheet is a passive response of the putative anchor region in vivo, which is controlled by the
charged residues. Such behavior could be influenced by local proton or ion gradients in the
cytosol. To test, if the emerged secondary structure was determined by its trajectories
extracted form the crystal structure, the protonation states during the runs were altered. The
simulations were started under acidic or basic conditions, respectively. After 10 ns, the
trajectories of the last calculated structure was extracted and used for the following run under
the next nominal pH condition for additional 10 ns. As described in figure 3-6, runs from
acidic to neutral to basic and vice versa were performed. There was no alteration in secondary
structure between the acidic and the neutral run. However, after changing the conditions from
neutral to basic, the anti-parallel beta-sheets emerged. The simulations which started with the
basic run delivered a similar result (figure 3-6B). During the first 10 ns the anti-parallel beta-
sheet was visible after ~3.5 ns and remained constant until the end of the basic run at 10 ns.
During the run under “neutral” conditions, the beta-sheet was available until ~18 ns and
vanished at the last 2 ns of that step. Under “acidic” conditions, there was also no beta-sheet
conformation detectable. This result, together with the determined secondary structure under
neutral conditions visualized under 3-3, indicates a bi-stable behavior for the peptide under
nominal “neutral” conditions. The anti-parallel as well as the unstructured configuration was
calculated. Further it suggests that the starting structure does not determine the secondary
structure. The anti-parallel beta-sheet formation depends on the amino acid composition.
Results
43
Figure 3-3: (A) Emergent secondary structure of vinculin´s C-terminal 21 residues under charge “basic” conditions according to DSSP. (B) Clustered conformations of 21 residue vinculin-tail peptide and putative lipid anchor. The fusion distance is a scale bar for measuring the distances between compact clusters. The labels on the x-axis represent the snapshots taken during the molecular dynamics simulation. The distinguished clusters are colored black and red, respectively. The representative structure of each cluster is colored in green. Note, that the asterisk marks the most compact cluster which is magnified. (C) The representative backbone geometry of the most compact cluster at time point 7100 ps.
Results
44
Figure 3-4: (A) Emergent secondary structure of vinculin´s C-terminal 21 residues under charge “neutral” conditions according to DSSP. (B) Clustered conformations of 21 residue vinculin-tail including the lipid anchor. Note, that the most compact cluster was magnified and labeled by an asterik. (C) The representative backbone geometry of the most compact cluster was at time point 8400 ps.
Results
45
Figure 3-5: (A) Emergent secondary structure of vinculin´s the C-terminal 21 residues under charge “acidic” conditions according to DSSP. (B) Clustered conformations of 21 residue vinculin-tail including the lipid anchor. Note, that the most compact cluster was magnified and labeled by an asterik. (C) The representative backbone geometry of the most compact cluster at time point 8400 ps
Results
46
Figure 3-6: MD-simulations of the 21 residue peptide, representing vinculin´s unstructured C-terminal arm. The nominal pH conditions were altered each 10ns from (A) “acidic” to “neutral” to “basic” or (B) from “basic” to “neutral” to “acidic”. The result implies that the secondary structure was not influenced by the starting structure.
3.2.2 MD Simulations of the vinculin-tail
The results determined in 3.2.1 suggest that the lipid anchor of vinculin is involved in beta-
sheet formation. All those simulations were performed out of the context of the vinculin-tail.
To test if the lipid anchor is also involved in beta-sheet formation, molecular dynamics
simulations were performed, using the whole vinculin-tail molecule (residues 890-1066). The
trajectories were also extracted from the crystal structure (PDB 1ST6) of chicken vinculin and
prepared as described under 2.2.4.1. The simulations were performed under charge neutral
conditions. Because of the emerged secondary structure under “neutral” conditions, which
reflects the cytosolic pH, the MD-simulation of the whole vinculin-tail was also performed
under those conditions. As described in figure 3-7, no beta-sheet emerged at the lipid anchor
region in the context of the whole vinculin-tail molecule. The C-terminal arm remained
unstructured during the whole simulation. This may imply that a conformational switch of the
vinculin-tail, induced by local pH shift or interaction with binding partners such as actin, is
needed to set the unstructured C-terminal arm free which could allow for anti-parallel beta-
sheet formation of the lipid anchor.
Results
47
Figure 3-7: Secondary structure of the vinculin-tail (residue 890-1066). The red square marks the unstructured C-terminal 21 residues (residues 1045-1066) including the lipid anchor (1052-1066). Note that no anti-parallel beta-sheet emerged after 150 ns.
3.3 CD-spectroscopy measurements of the C-terminal arm The MD-Simulations of the vinculin peptide, representing its C-terminal arm, revealed an
involvement of the lipid anchor in beta-sheet formation under nominal “basic” and “neutral”
conditions (refer to 3.2). Furthermore, it was demonstrated by DSC measurements that this
peptide has the ability to insert into lipid vesicles (refer to 3.1). To proof the in silico
determined secondary structure in the presence and absence of lipid vesicles CD-
Results
48
spectroscopic measurements were performed. As described under 2.2.3.5, SUVs consisting of
DMPC/ DMPG at a molar ratio of 70:30 were used in these experiments. At this composition,
the best insertion behavior was determined according to DSC measurements (refer to 3.1).
With respect to the cytosolic environment, all measurements were performed at physiological
pH. The CD-spectra of the C-terminal peptide in the presence/ absence of lipid vesicles were
plotted in figure 3-8. Measuring between 180 and 260 nm, the spectrum of pure peptide
showed a minimum at 200 nm, whilst in the presence of SUVs this minimum was shifted to
223 nm. This observation suggests a clear change in conformation of the peptide. Analyses of
the spectral data using algorithms (CDSSTR and CONTINLL) that employ different
approaches for deconvoluting the CD data from a database of secondary structure
contributions, show similar structural elements within the vinculin C-terminal peptide. As
described in figure 3-8B, CDSSTR shows 25 % beta-strand 1, 13 % beta-strand 2, 43 %
random coil (NRMSD < 0.016) and CONTINLL 21.9 % beta-strand 1, 13 % beta-strand 2,
42.6 % random coil (NRMSD < 0.134).
Figure 3-8: Results from CD-spectroscopic measurements of the C-terminal peptide in the presence/ absence of SUVs. The CD spectra of 70 µM peptide were obtained in 10mM sodium phosphate at pH 7.4 and 30 °C (n = 3). (A) The plot of the mean residue molar ellipticity, Θ (deg cm2dmol-1) in the presence and absence of DMPC/ DMPG lipid vesicles. Secondary structure prediction according to the algorithms provided by Dichroweb in the (B) absence and (C) presence of lipid vesicles. Note, that the result is in good agreement with the calculated secondary structure using GROMACS.
Results
49
These analyses indicate a possible anti-parallel beta-sheet conformation. It underlines that the
CD spectrum of the peptide allows an interpretation which agrees with the representative at
8400 ps derived from the MD simulation under “neutral” conditions. In the presence of SUVs,
CD spectroscopy of the vinculin peptide provided similar results (figure 3-8C). In comparison
to the spectra in the absence of lipid vesicles, the two algorithms indicate also two beta-
strands and unstructured parts at a similar distribution (NRMSD< 0.05) for both CDSSTR and
CONTINLL. However, the difference in spectral characteristics of the spectra for the peptide
in the presence/ absence of lipid vesicles suggests distinguishable beta-sheet conformations.
3.4 Solid state NMR measurements of the C-terminal arm The results of the MD-simulations together with the CD-spectroscopic measurements in
presence and absence of lipid vesicles indicate that the lipid anchor is involved in an anti-
parallel beta-sheet formation whilst interacting with lipid membranes. To proof the orientation
of this interaction, solid state NMR measurements were performed as described under 2.2.3.6.
The 15N- label was placed in the leucine residue on position 11 of the C-terminal arm peptide
(IKIRTDAGFTLRWVRKTPWYQ), which was part of the second beta-strand. One single
peak at 100 ppm or 200 ppm would suggest a distinct conformation of the peptide during lipid
membrane interaction. As described in figure 3-9, a plateau around 200 ppm followed by one
prominent single peak at ~100 ppm appeared. This implies a random distribution with a
tendency to one certain conformation during interaction (in detail of the labeled part of the
peptide) with the lipid membrane. The shift to ~100 ppm together with the proven secondary
structure of the beta-sheet (refer to 3.5 and 3.6) implies a surface seeking interaction of the
peptide. The spectra of the lipids (figure 3-9B) demonstrated a well oriented lipid bilayer into
which the peptide inserted. Additional measurements are necessary to elucidate the lipid
interaction behavior and the insertion angle of the peptide in more detail.
Results
50
Figure 3-9: Result of the NMR measurement of vinculin´s unstructured C-terminal arm. (A) Proton-decoupled 15N solid-state NMR spectra of 10mg peptide reconstituted into 100 mg of 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC)/ palmitoyl-2-oleoyl-sn-glycero-3-phosphoglycine (POPG) membranes at a molar ratio of 70 : 30. (B) The proton-decoupled 31P solid-state NMR spectrum of the sample is shown in panel A. The 31P NMR line shape represents the distribution of alignments of the phospholipid head group in the sample.
II. In vivo 3.5 Cloning and expression of the different vinculin constructs
3.5.1 Cloning and expression of EGFP linked vinculin and vinculinΔC
The structural and functional role of vinculin´s lipid anchor in the regulation of adhesion sites
was studied using recombinant vinculin proteins. Eukaryotic expression constructs were used
for biochemical and cellular characterization of the protein. EGFP linked full length vinculin
and vinculinΔC was synthesized as described under 2.2.1.13. To test if the EGFP labeled
constructs were localized at the FA, MEF-vin(-/-) cells were transfected with the vinculin and
vinculinΔC constructs (Refer to chapter 2.2.2.2), reseeded on coverslips (refer to 2.2.2.3) and
fixed (refer to 2.2.2.4) for microscopic analysis. Furthermore, the lysates of the transfected
cells were used for Western blot analysis (Refer to chapter 2.2.3.9.). Figure 3-10 shows the
images of the MEF-wt (wildtype), MEF-vin(-/-) (vinculin knock out), MEF-resc (rescue) and
MEF-vin C (transfected with vinculin lacking the lipid anchor; residues 1-1052) cells. The
Results
51
focal adhesions (green) were determined in MEF-wt and MEF-vin(-/-) cells using antibodies
against vinculin. In MEF-resc and MEF-vin C cells the EGFP molecule connected to the
different vinculin peptides was used. As observed in the pictures, vinculin is localized in the
focal adhesions. The focal adhesions are connected to the actin stress-fibers colored in red.
According to the Western Blot analysis (Figure 3-10B) vinculin and vinculinΔC linked to
EGFP were expressed in the cells. There are no additional bands below 120 kDa visible,
suggesting there was no degradation of the vinculin peptides.
Figure 3-10: Expression of vinculin and vinculinΔC in the different MEF cell lines determined by (A) Immunofluorescence and (B) Western blot analysis. (A) The FA were stained with a monoclonal antibody against vinculin in MEF-wt and MEF-vin(-/-). For visualization of the vinculin constructs in MEF-resc and MEF-vinΔC, the EGFP label was used. The actin cytoskeleton was stained with TRITC-Phalloidin in each cell line. The scale bar represents 50 µm. The Inset shows FA in MEF-vin(-/-) stained with paxillin (B) Western Blot analysis show the cell lysates of MEF-wt, MEF-vin(-/-), MEF-resc and MEF-vinΔC cells. There was no vinculin in the MEF vin(-/-) cells detected. Note, that the EGFP tag on the different vinculin constructs in MEF-resc and MEF-vinΔC cells caused a shift of the molecule in the direction to higher molecular weight.
Results
52
3.5.2 Cloning and expression of EGFP-linked vinculin-tail and vinculin-tailΔC
F9 Vinculin (-/-) carcinoma cells showed a decrease in stiffness and develop less force in
comparison to their wildtype cells [65; 69]. Furthermore, it was reported that the
retransfection of the vinculin-tail domain is sufficient to fully restore the wildtype phenotype,
suggesting that the interaction between the actin cytoskeleton and vinculin is responsible for
those effects [65]. To test the influence of the lipid anchor to cell mechanical behavior in
absence of the vinculin-head, an EGFP linked vinculin-tail (vtail, residues 858-1066) and
vinculin-tailΔC (vtailΔC, residue vtail 858-1052) construct was prepared as described under
2.2.1.13, using the primer pair 2b/ 2f or 2e/ 2f, listed under 2.1.3. MEF-vin(-/-) cells
transfected with EGFP linked vtail and vtailΔC were used to obtain the difference to cell
mechanics of this two constructs (refer to chapter 2.2.2.2). As described in figure 3-11, both
constructs were localized at the FA. However, both vinculin-tail peptides also masked the
actin stress-fibers. The cytosolic background was higher in comparison to MEF-vin(-/-) cells
transfected with the full length vinculin constructs (see figure 3-11). The cells expressing the
vtail constructs did not spread well, suggesting that the vtail and vtailΔC constructs show an
impaired FA localization. An expression analysis of the different vtail constructs via Western
Blot analysis was not possible with the used antibody because the epitope of recognition is
localized in the vinculin-head region.
Figure 3-11: Expression of EGFP linked vinculin-tail and vinculin-tailΔC in MEF-vin(-/-) cells determined by fluorescence microscopy. The scale bar represents 20 µm.
3.5.3 Mutagenesis of vinculin´s src dependent phosphorylation site Y1065F
The phosphorylation of vinculin also influences the mechanical properties of the cells. It was
reported that the mutation of the c-src dependent phosphorylation sites on vinculin results in
impaired cell spreading and migration [71-72]. One of this phosphorylation sites, Tyrosine
1065, is localized at the lipid anchor. To determine the influence of this site to cell mechanics,
the tyrosine residue on position 1065 was mutated to a phenylalanine by site-directed
Results
53
mutagenesis (refer to 2.2.1.8). This construct was transfected into MEF-vin(-/-) cells and
transferred onto FN-coated glass coverslips (refer to chapter 2.2.2.3). After fixation the actin
cytoskeleton was stained with TRITC-Phalloidin. According to figure 3-12 the
vinculinY1065F construct was properly localized to the FA. Further, no visible difference in
focal adhesion size was determined in comparison to non-mutated vinculin.
Figure 3-12: Expression of vinculin and vinculinY1065F in MEF-vin(-/-) cell lines (green). The actin cytoskeleton was stained using TRITC-Phalloidin (red). The scale bar represents 20 µm.
3.6 Magnetic tweezer measurements Control of cell mechanics and shape is crucial for many cellular functions, including growth,
differentiation, migration, and gene expression. The mechanical properties of the cell, in turn,
are established through a balance of mechanical forces. Tensions, generated by actin filaments
are counteracted by compression resistant extracellular matrix (ECM) anchors as well as
internal microtubule structures [5-6]. To measure the intracellular prestress a magnetic
tweezer was used for the quantification of mechanical properties of individual living cells [69;
103]. The magnetic tweezer measurements were performed as described in chapter 2.2.5.1. In
brief, transfected and non-transfected cells were incubated with fibronectin-coated para-
magnetic beads 30 min prior to measurements. During the measurements, a solenoid with a
needle-shaped core was placed in front of a FN-coated bead connected to a single cell. After
the application of a magnetic field, the bead moved towards the needle. The displacement of
the bead (d) follows a power law
btta )/(d 0= (15)
where parameter a (units of µm/nN) characterizes the elastic cell properties and corresponds
to a compliance whilst the exponent b reflects the viscoelasticity. The inverse of compliance
is defined as stiffness 1/a (units of nN/µm).
Results
54
3.6.1 Magnetic tweezer measurements of MEF-resc and MEF-vinΔC cells
Cell attachment to the ECM via focal adhesions is prerequisite for cell survival and mainly
driven by transmembrane receptors, like intergin [22]. The co-localization of the focal
adhesions and the beads were tested by confocal microscopy. MEF-vin(-/-) cells transfected
with EGFP linked vinculin or vinculinΔC, were seeded on coverslips (refer to 2.2.2.3),
incubated with FN-coated beads and fixed (refer to 2.2.2.4). As control, the focal adhesions of
MEF-wt and MEF-vin(-/-) cells, seeded and fixed on cover slips, were stained using an
antibody against the FA protein paxillin (refer to 2.2.2.5). The images in figure 3-13 show
MEF-wt, MEF-vin(-/-), MEF-resc and MEF-vin C cells.
Figure 3-13: Co-localization of the auto-fluorescent beads (red) with the focal adhesions (green). Transfected and non-transfected MEF vin(-/-) cells were seeded on glass slides. The cells were allowed to adhere overnight. Prior to fixation, the cells were incubated with FN-coated beads for 1 h. The FAs in MEF-wt and MEF-vin(-/-) cells were stained with antibodies against paxillin; in MEF-vin(-/-) cells transfected with vinculin (MEF-resc) and vinculinΔC (MEF-vin C) focal adhesion were determined through visualization of the N-terminal EGFP label. The arrows mark the FN-coated beads which are in close proximity to the focal adhesions.
Results
55
The focal adhesions (green) were determined in MEF-wt and MEF-vin(-/-) cells using
antibodies against paxillin, and in MEF-resc and MEF-vin C cells using the EGFP molecule
connected to the different vinculin peptides. The position of the 4.5 µm FN-coated beads (red,
arrow) was determined using its auto-fluorescent properties. In close proximity to the beads,
an accumulation of the FA components such as paxillin or vinculin was detected, suggesting
that beads were linked to focal contacts. These pictures represent this common phenomenon.
Note, that there was no visual difference in FA size in all of these cells. Cells expressing
vinculin lacking the lipid anchor region showed a reduced FA turnover rate and impaired cell
migration [81]. This implies that the lipid anchor influences cell mechanical behavior. To
determine the influence of vinculin´s lipid anchor on cell stiffness MEF-wt, MEF-vin(-/-),
MEF-resc and MEF-vinΔC were measured using the magnetic tweezer device. Transfected
and non-transfected cells were measured as described under 2.2.5.1. The results are shown in
figure 3-14A, where we applied forces of 1 nN to FN-coated beads attached to MEF-wt and
MEF-resc cells. These cells indicated similar stiffness, whilst MEF-vin(-/-) and MEF-vin C
cells showed significantly reduced stiffness values of ~ 48 % and ~26 %, respectively,
compared to MEF-wt cells. This suggests that the lipid anchor is important for cell stiffness.
To test, whether the binding strength of the different MEF-cells to fibronection-coated
surfaces holds over a wide range of forces, the externally applied force was increased over 10
s in a staircase-like manner from 0.5 – 10 nN. With increasing forces (0.5 – 10 nN), the bead
detachment also increased (figure 3-14B). At 10 nN force, ~20 % and ~14 % beads detached
from MEF-wt and MEF-resc, respectively, whilst 37 % of the beads detached during force
application on MEF-vin(-/-) cells. MEF cells transfected with vinculin C showed at low
forces (0.5 - 5 nN) similar behavior to MEF-wt and rescue cells. Only at higher (5 – 10 nN)
force nearly the same level of beads detached in comparison to MEF-vin(-/-) cells. The
fraction of beads that detached at a given force level can be regarded as a direct measure of
the adhesion strength (yielding force) of the force-transmitting elements and bonds between
the fibronectin-coated beads, integrins, focal adhesion proteins and the cytoskeleton. These
results may imply that the lipid anchor of vinculin influences the adhesion strength between
the cytoskeleton and ECM-coated surfaces.
Results
56
Figure 3-14: Result of magnetic tweezer measurements. A) Stiffness (geometric mean) of the transfected and non-transfected MEF cells determined at 1 nN force. The error bars represent the standard error (SEM). B) Percentage of detached beads vs. forces in MEF-wt and vinculin mutant cells averaged for all experiments. Besides the increased bead detachment level at high forces, the differential stiffness of all cell
lines increased also with the rising force. In figure 3-15 the stiffness of the different MEF-cell
lines was plotted against the applied force. The stiffness calculation in figure 3-15A includes
the displacements of the detached beads. In figure 3-15B, the stiffness was calculated only
with the beads that remained attached until the end of the measurement at 10 nN force. MEF-
wt and MEF-resc cells displayed at all force steps significantly higher stiffness values in
comparison to MEF-vin(-/-) cells, independent from the level of disrupted beads. In contrast,
the stiffness values for MEF-vin C cells were over a wide range significantly less stiff in
comparison to MEF-wt and MEF-resc performing the calculation with all measured beads
(figure 3-15A). Only at high forces (6 – 10 nN) they reached almost wildtype and rescue
Results
57
level. Calculating the stiffness of MEF-vin C cells excluding the detached beads (figure 3-
15B) demonstrated that MEF-vin C cells showed similar stiffness levels than MEF-wt and
MEF-resc cells over the whole force range. In contrast, the stiffness of MEF-vin(-/-) cells was
always ~50 % decreased in comparison to MEF-wt and rescue, independent from the
disrupted beads. This suggests that the reduced binding strength in vinculinΔC cells is
influencing also its stiffness.
Figure 3-15: Differential cell stiffness of transfected and non-transfected MEF cells at given forces between 0.5 to 10 nN visualized in a log-log plot (p<0.05). The cell stiffness was determined including (A) or excluding (B) the detached beads. The stiffness increased with force nearly parallel for MEF-wt, MEF-resc, MEF-vin(-/-) cell lines independent from the detachment level. However, calculating the stiffness of MEF-vinΔC cells, including the disrupted beads (A), the values were significantly different to MEF-wt and MEF-resc for forces between 0.5 – 4 nN. The application of higher forces resulted in stiffness values near to MEF-wt and -resc cells. Excluding the detached beads (B) resulted in similar stiffness than MEF-wt and MEF-resc cells. The error bars represent the standard error (SEM). To determine the viscoelasticity of the transfected and non-transfected cells, the power-law
exponent b was plotted against the applied force (figure 3-16). The exponent b determines the
visco-elastic behavior of the cells [65; 103]. The values are between 0 and 1, whereas 0
reflects elastic and 1 viscous behavior. In figure 3-16A the b-Factor was calculated using all
Results
58
measured beads, whilst in figure 3-16B the disrupted ones were excluded. The exponent b of
the MEF-vin(-/-) cells was significantly higher in comparison to MEF-wt, MEF-resc and
MEF-vinΔC over the applied force range (force 1-10 nN), independent from the level of
disrupted beads. This indicates a more viscous behavior for MEF-vin(-/-) cells in comparison
to MEF-wt and MEF-resc cells. No significant difference was detected between MEF-wt,
MEF-resc and MEF-vinΔC cells, suggesting that the lipid anchor is not affecting the
viscoelasticity of the cell.
Figure 3-16: Geometric mean of the power-law exponent b over a range of applied forces from 0.5-10 nN, calculated with all (A) and without the disrupted beads (B). There was no significant difference between MEF-wt, MEF-resc and MEF-vinΔC cells. The error bars represent the standard error (SEM).
3.6.2 Creep measurements of MEF-vtail and MEF-vtailΔC cells
F9 mouse carcinoma cells expressing only the vinculin-tail showed nearly the same
mechanical properties as the F9 wildtype or rescue cells [65]. It is assumed that this is due to
vinculin´s ability to interact with the actin cytoskeleton [65]. Generally, the access of the
Results
59
vinculin-tail is controlled by the vinculin-head domain which masks the cryptic binding-sites
for actin, paxillin and phospholipids localized at the tail [52]. Talin- and also lipid interaction
displace the vinculin-head domain from the tail and free those interacting sites [20]. To
reveal, whether the influence on cell mechanics of the lipid anchor is due to the activation or
the anchorage of vinculin to the membrane itself, magnetic tweezer measurements (refer to
2.2.5.1) were also performed using MEF-vtail (residue 858-1066) and MEF-vtailΔC (residue
858 – 1052) cells. The results are plotted in figure 3-17. Forces of 1 nN were applied to FN-
coated beads attached to the MEF cells, transfected with the different EGFP-linked vinculin-
tail constructs. MEF-vtail and MEF-vtail C cells showed a stiffness reduction to
approximately 40 % in comparison to MEF-wt. There was no difference in stiffness between
MEF-vtail and MEF-vtail C, suggesting that the lipid anchor only influence stiffness in cells
expressing full length vinculin constructs. Furthermore, measuring the binding strength
between the beads and MEF-vtail and MEF-vtailΔC cells revealed higher detachment values
in comparison to MEF-vin(-/-) cells (figure 3-17; B). At 10 nN force, ~41 % and ~52 % of
beads detached from MEF-vtailΔC and MEF-vtail, respectively. In contrast, only 37 % of the
beads detached from MEF-vin(-/-) cells. The account of detached beads did not move
between the values determined by MEF-vin(-/-) and MEF-wt, suggesting that the expression
of the EGFP linked vinculin-tail variants negatively affects cell mechanical behavior.
Results
60
Figure 3-17: Magnetic tweezer measurements of MEF-vtail and MEF-vtailΔC in comparison to MEF-wt and MEF-vin(-/-) cells. A) Stiffness (geometric mean) of transfected and non-transfected MEF cells determined at 1nN force. There was no detectable difference between MEF-vtail and MEF-vtailΔC. The error bars represent the standard error (SEM). B) Percentage of detached beads vs. forces, averaged for all experiments. Note, that the detachement levels of MEF-vtail and MEF-vtailΔC were not in between those values, determined by MEF-wt and MEF-vin(-/-) cells. In figure 3-18 the stiffness of MEF-wt, MEF-vin(-/-), MEF-vtail and MEF-vtailΔC was
plotted against the applied force ramp. Figure 3-18A represents the stiffness calculation
including the displacements of the detached beads. The stiffness values of the cells transfected
with the different vinculin-tail constructs were between the values of MEF-wt and MEF-vin(-
/-) cells. No difference in stiffness was determined for MEF-vtailΔC and MEF-vtail over the
whole force range. In figure 3-18B, the stiffness was calculated only with the beads that
Results
61
remained attached until the end of the measurement. There was also no difference in stiffness
between MEF-vtailΔC and MEF-vtail. Furthermore, MEF-vtailΔC and MEF-vtail were as
stiff as MEF-wt over the whole force range, suggesting the vinculin-tail hosts the mechano-
coupling properties.
Figure 3-18: Differential cell stiffness of MEF-vin(-/-) cells transfected with vtail and vtailΔC at given forces between 0.5 to 10 nN visualized in a log-log plot. The cell stiffness was determined (A) with all (A) as well as without the disrupted beads (B). The stiffness increased with force nearly parallel for all measured cell lines independent from the bead detachement level and the values. The error bars represent the standard error (SEM). In figure 3-19, the viscoelasticity (determined by the power-law exponent b) of cells
expressing the vinculin-tail constructs, was plotted against the applied force. Including all
beads into the calculation, the power-law exponent b of the MEF-vtail and MEF-vtailΔC cells
jumps between the values of MEF-wt and MEF-vin(-/-). The evaluation of the b-values with
the beads that remained attached until the end of the measurements delivered a clearer picture
Results
62
(figure 3-19B). The MEF-vtail and MEF-vtailΔC are almost as elastic as the MEF-wt cells.
No difference between MEF-vtail and MEF-vtailΔC was detected. This could be seen as
additional evidence that the lipid anchor is not affecting the viscoelasticity of the cell.
Figure 3-19: The mean of the power-law exponent b over a range of applied force from 0.5-10 nN, determined including (A) or excluding (B) the detached beads. There is no significant difference between MEF-wt, MEF-vtail and MEF-vtailΔC, suggesting no effect of the lipid anchor due to the cellular viscoelastic behavior. The error bars represent the standard error (SEM).
During the creep measurements, no significant differences in stiffness, binding strength and
viscoelasticity between MEF-vtail and MEF-vtailΔC could be detected, independent if the
calculation was performed with all beads or only with the beads which remained attached
until the end of the measurement at 10 nN. The result suggests that the lipid anchor influence
stiffness only in cells expressing full length vinculin constructs. This could be indicative for
Results
63
an involvement of the lipid anchor in regulating vinculin´s head/ tail interaction. In the
following experiments only full length vinculin constructs were used.
3.7 Determination of the spreading area Cell stiffness and force generation are important for cell adhesion and migration [18]. To test
the influence of the lipid anchor on cell mechanic properties, the spreading area was
determined. As described in chapter 2.2.2.6, transfected and non-transfected cells were seeded
on FN-coated coverslips. After 1h incubation, the samples were fixed and images were taken
(refer to chapter 2.2.2.6). As demonstrated in figure 3-20, MEF-vin(-/-) cells were half the
size of MEF-wt. In contrast, Saunders and co-workers demonstrated that MEF-wt cells are
double the size in comparison to MEF-vin(-/-) cells [81]. No significant difference was
detected between MEF-resc and MEF-vinΔC cells. Furthermore, the retransformation of
vinculin in MEF-vin(-/-) cells did not fully restore the wildtype phenotype, suggesting the
vinculin knock out caused subsequent alterations which determines the spreading area of the
cell. Thus the data are not consistent, meaning the spreading area is no criteria for evaluating
any difference between MEF-wt, MEF-resc and MEF-vinΔC cells.
n=108
n=96
n=566
n=498
0
500
1000
1500
2000
2500
3000
3500
4000
MEF-wt MEF-vin(-/-) MEF-resc MEF-vin∆C
Spre
adin
g ar
ea (µ
m2 )
Figure 3-20: Spreading area of transfected and non-tranfected MEF cells. Note, that the data were inconsistent because they were not in the given parameters, determined by the MEF-wt and MEF-vin(-/-) cells. The error bars represent the standard error (SEM).
3.8 Determination of the FA per cell The FAs are the linkers between the ECM and the actin cytoskeleton. The connection to the
extracellular matrix determines cell size and shape as well as controls its survival. It was
reported that the vinculin molecule influences focal adhesion enlargement and its maturation
[108]. It is also known that the number of focal adhesions depends on the internal tension of
the cell [109]. Because of the inconsistent spreading area (see also chapter 3.7), the number of
focal adhesions in the different MEF cell lines was determined using cells of a similar
Results
64
spreading area (Figure 3-21). MEF-wt and MEF-resc cells showed a similar number of focal
adhesions, whilst MEF-vin(-/-) cells transfected with vinculinΔC displayed 23 % and MEF-
vin(-/-) cells showed 40 % less focal adhesions compared to MEF wildtype and rescue cells.
This result suggests an involvement of the lipid anchor in focal adhesion formation.
Figure 3-21: Determination of focal adhesions (FA) per cell line. (A) MEF-wt and MEF-resc show a similar number of FA. MEF-/- cells transfected with vinculin∆C show sligthly more FAs in comparison to wt and rescue cells. To stain the FA in MEF-wt and MEF-vin(-/-) cells, an antibody against paxillin (green) was used. In MEF-resc and MEF-vinΔC cells, the EGFP-molecule linked to the vinculin proteins was used (green). (B) Numbers of the FA plotted for each cell line. The spreading area of the cells used for FA determination was ~4000 µm2 in all cell lines (inset). The error bars represent the standard error (SEM).
Results
65
3.9 FRAP measurements MEF-cells expressing vinculinΔC showed a reduced focal adhesion turnover rate whilst
there was no detectable difference in the exchange rate of vinculin and vinculinΔC molecules
localized in FA, judged by FRAP measurements [81]. However, in this article they did not
report wether they determined the recovery half time in the presence or absence of
endogenous vinculin.
Figure 3-22: Turn over rate of EGFP-vinculin and vinculinΔC transiently expressed in MEF-vin(-/-) cells. The red circle marks the measured FA before and after bleaching. There was no significant difference in recovery half time (t ½) between MEF-resc and MEF-vinΔC cells. The error bars represent the standard error (SEM).
Results
66
To test the exchange rates of EGFP-linked vinculin and vinculinΔC molecules in individual
adhesion sites expressed in MEF-vin(-/-) cells, fluorescence recovery after photo bleaching
(FRAP) experiments were carried out. The technique calculates the rate of re-incorporation of
GFP-tagged vinculin variants after bleaching them with 100% laser intensity. Transfected
MEF-vin(-/-) cells expressing GFP vinculin and vinculinΔC were seeded on glass bottom
wells and the measurements were performed as described in 2.2.2.7. As described in figure 3-
22, individual adhesions were bleached by increasing the laser intensity and subjected to time
lapse video microscopy. The halftime of fluorescence recovery after photobleaching was
about 96,91s for vinculin and 119,86s for vinculinΔC molecules. According to the standard
error, there was no detectable difference in exchange rates of the protein in individual
adhesion sites, suggesting the anchor does not influence the exchange rate of individual
vinculin molecules.
3.10 2D-traction microscopy measurements During their life cells exert tractions on their surroundings including contraction, spreading,
crawling as well as invasion. These functions are associated with complex mechanical
interactions between the extracellular substrates, focal adhesion molecules such as vinculin,
cytoskeletal elements, and molecular motors. Dembo and co-workers recently demonstrated
that the traction field of a cell exerted on its surroundings can be mapped from knowledge of
the displacement field in a flexible substrate on which cells are adherent [110]. The
measurement of the displacement field is accomplished by tracking small beads, embedded on
the surface of polyacrylamide gels of a known elasticity module. The raw displacements
themselves can be seen as a qualitative map of the local tractions, generated by a single cell.
3.10.1 Strain energy measurements of MEF-vinΔC
The distribution, the size and the numbers of FAs are dependent on internal and external
forces [25; 29]. The contractile forces of the actomyosin motor activity, transferred via the
focal adhesions to the ECM-coated PAA gel, was tested in these experiments using various
MEF cell lines. Figure 3-23A shows images of MEF-wt, MEF-vin(-/-), MEF-resc and MEF-
vinΔC together with the traction-fields generated by those cells. MEF-vin(-/-) cells displayed
about 80 % and MEF-vin(-/-) cells transfected with the vinculinΔC mutant showed about 41
% less traction force during adhesion compared to the MEF-wt and rescue cells (figure 23B).
These results suggest that vinculin´s lipid anchor is important for generating the appropriate
force during cell adhesion.
Results
67
Figure 3-23: Results of the 2D-traction microscopy measurements for MEF-wt, MEF-vin(-/-), MEF-resc and MEF-vinΔC. (A) Pictures of the traction maps and the corresponding cells, generating the deformations. The pseudo-color range in the pictures of the strain fields represents the different strain field values. To visualize the MEF-resc and MEF-vinΔC, the EGFP label was used. The scale bar represents 50 µm. (B) Plot of the elastic strain energy (mean) stored in the extracellular matrix due to cell tractions. The error bars represent the standard error (SEM).
3.10.2 Strain energy measurements of different vinculin mutants
Previously, it was demonstrated by Chandrasekar et al. that the mutation of six surface
exposed basic residues on vinculin-tail to glutamine (Q) in helix 3 ( H3; K952, K956, R963,
R966) and the C-terminal lipid anchor (CT; R1060, K1061) result in impaired lipid membrane
interaction [82]. To test, whether the difference in tractions is due to the lipid binding of
vinculin in general, we performed additional 2D-traction microscopic measurements with
those vinculin mutants deficient in lipid binding. MEF-vin(-/-) cells were transfected (refer to
chapter 2.2.2.2) with EGFP-linked vinculin-H3 (K952, K956, R963, R966 to Q), vinculin-CT
(R1060, K1061 to Q) and vinculin-LD (K952, K956, R963, R966, R1060, K1061 to Q; equals
CT + H3) constructs. As described in figure 3-24A, all constructs were localized in the focal
adhesions. There was no detectable difference in FA size and distribution. The 2D traction
measurements were performed as described in chapter 2.2.5.2. The results from these
experiments were plotted as relative strain energy compared to MEF-resc cells (figure 3-24B).
Results
68
MEF-vin(-/-) cells transfected with the lipid binding deficiency variant of vinculin displaying
all six point mutations (MEF-LD) showed a reduction in strain energy of about 50 %
compared to MEF-resc cells. MEF-CT cells, expressing a vinculin variant with two point-
mutations in the lipid anchor region which also impairs lipid binding displayed the same
reduced strain energy level than MEF-LD. However, MEF-vin(-/-) cells carrying vinculin
with the mutated H3 domain, which also influence the lipid binding [82], showed a similar
strain energy level as MEF-resc cells, suggesting that only the change of vinculin´s lipid
anchor localized at the C-terminus affects force generation in MEF cells.
Figure 3-24: Results of the strain energy measurements of MEF-LD, MEF-CT and MEF-H3 cells. (A) Pictures of MEF-vin(-/-) cells transfected with EGFP labeled vinculin-LD (MEF-LD), vinculin-CT (MEF-CT) and vincunlin-H3 (MEF-H3). The actin cytoskeleton was stained with TRITC-Phalloidin in each cell line. The scale bar represents 20 µm. (B) Relative strain energy (mean) stored in the PAA gel, coated with the extracellular matrix protein collagen. The energy values of LD (K952, K956, R963, R966 and R1060, K1061 transferred to Q), CT (R1060, K1061 transferred to Q) and H3 (K952, K956, R963, R966 transferred to Q) were correlated to the rescue (resc) values. The absolute strain energy values of MEF-LD, -H3 and CT were ~1 pJ. The energy values of Y1065F were ~10 pJ (Note, the higher value was due to different gel coating). The error bars represent the standard error (SEM).
Results
69
Furthermore, a src-dependent phosphorylation site is also localized at the lipid anchor [71-
72]. The mutation of this tyrosine phosphorylation site on position 1065 to phenylalanine
(Y1065F) results in impaired cell spreading and migration. This suggests that the src
phosphorylation affects cell mechanical behavior. As demonstrated in Figure 3-24, the
vinculinY1065F mutant also showed a decreased force generation in comparison to the rescue
cells. The lipid binding abilities of vinculinY1065F are not impaired. This implies that the
interaction of the lipid anchor with the cell membrane influences cell mechanical behavior via
src-phosphorylation.
In summary, vinculin´s membrane binding of the lipid anchor influences cell stiffness and is
necessary for traction generation. 2D-traction measurements with MEF-Y1065F as well as
MEF-CT imply that the anchorage of the C-terminal arm to the cell membrane provides
vinculin for src phosphorylation. This in turn may alter the affinity of vinculin for other
binding partners such as the actin crosslinker Arp2/3 [72].
Discussion
70
4 Discussion Signal transduction, vesicle trafficking, retroviral assembly, focal adhesion formation and
other central biological processes directly involve membrane-embedded protein complexes.
Over the last decades, several structural elements have been identified that provide proteins
for transiently lipid membrane interaction [38; 111-112]. There are proteins such as spectrin,
an actin associated protein, which show pleckstrin homology (PH) domains. This binding
motif recognizes different phosphatidylinositol (PIP) lipids localized in cell membranes [111].
Another PIP binding motif is the FERM domain, named after the four proteins in which this
domain was originally described: F for Band 4.1, E for Ezrin, R for Radixin, M for Moesin
[112]. Post-translational modifications, such as myristylation or palmitoylation, may also play
critical roles in regulating membrane associations [113]. There are also secondary structure
elements such as amphiphatic helices and hydrophobic beta-sheets that interact transiently
with lipid membranes [38; 75; 80; 112]. The application of biophysical techniques including
Fourier-transformed infrared spectroscopy (FTIR), neutron reflection, electron spin resonance
(ESR), nuclear magnetic resonance (NMR) and X-ray crystallography has been helpful in
characterizing these protein-membrane interaction in vitro [114]. Unfortunately, the
mechanism and structural consequences of membrane association in cells are still poorly
understood [38; 112].
The cytoskeleton and its associated proteins such as spectrin, alpha-actinin, Arp2/3, CapZ,
talin and vinculin are important ubiquitous cellular components that determine cell shape,
locomotion and adhesion [38; 75; 112]. To influence these processes, the cytoskeleton must
be connected reversibly to transmembrane receptors in a manner that is regulated by
signalling events. In the most cases, these receptors are linked via adaptor proteins, such as
talin, alpha-actinin and vinculin, to the cytoskeleton [19-20]. A variety of these cytoskeletal
interacting proteins have been shown to associate with, and in some cases insert into, lipid
bilayers [38; 112]. Indeed, in vitro evidence for specific membrane interactions has been
provided for several proteins and their interaction partners [38; 106; 112; 115-118]. In a few
studies, the existence of such interactions was also shown in intact cells [112; 119].
Experiments using a hydrophobic photolabel localized within the membrane of intact chicken
fibroblasts demonstrated the existence of such an insertion potenial for vinculin [119]. Further
experiments identified three lipid binding regions on the 30 kDa tail domain of vinculin [76-
78]. Whilst residues 935–978 and 1020–1040 were identified as two amphiphatic alpha-
helices [76-77], residues 1052–1066 (so-called lipid anchor) are involved in beta-sheet
formation [80]. Recent studies have shown that adhesion site dynamics and turnover as well
Discussion
71
as cell motility are directly affected by vinculin-tail binding to phospholipids [78; 81-82].
Two different strategies generating lipid-binding deficient vinculin mutants were examined:
one approach was based on a variant lacking the last 15 residues of vinculin’s C-terminus [78;
81]. In another experiment six point-mutations on surface-exposed basic residues (K952,
K956, R963, R966 and RK1060/61 to Q) were introduced to achieve similar impaired lipid
membrane binding [82]. However, the in vivo relevance of vinculin´s membrane interaction is
still far from being resolved [38; 112]. It was demonstrated that the interaction of vinculin
with acidic phospholipids regulates focal adhesion disassembly [82]. Other scientists propose
membrane binding triggers vinculin´s activation and FA recruitment [59-60; 74]. However,
this work revealed clear evidence for the involvement of vinculin´s lipid anchor in the
regulation of cell mechanical properties.
4.1 Vinculin´s lipid anchor revealed membrane insertion potential It was reported that expressed vinculin deficient in lipid binding still locates at focal
adhesions, but that the FA turnover was impaired [81-82]. Moreover, the same studies showed
that binding of the vinculin-tail to lipid membranes is dependent on the capacity of vinculin to
interact with acidic phospholipids, such as phosphatidylserine (PS) and phosphatidylinositol-
4,5-bisphosphate (PIP2). Work, focussing on the possible involvement of acidic phospholipids
in adhesion site turnover left the question unanswered of whether or not the vinculin lipid
anchor interacts directly with phospholipid membranes. To this end, an investigation into the
function of the C-terminal arm (residues 1045-1066), which includes the lipid anchor, was
required as many studies had previously indicated that the lipid anchor of the vinculin-tail
interacted with phospholipids such as phosphatidylcholine, phosphatidyl-serine as well as
PIP2 [78; 81]. To gain information about the thermodynamic behavior of artificial lipids in the
presence/ absence of the 21-residue C-terminal peptide of the tail domain, we applied
differential scanning calorimetry (DSC). Using multilemellar lipid vesicles (MLVs) allowed
the detection of subtle perturbations during the lipid melting process and gave direct evidence
of peptide– lipid interaction. As described in chapter 3.1.1, the C-terminal peptide of vinculin
showed membrane insertion behavior. Since membrane–protein interactions are known to
depend on the negative surface charge of membranes, we increased the charge of lipid
vesicles to mimic the inner leaflet of the cell membrane. The results showed that the phase
transition enthalpy increased, suggesting a reduced binding of vinculin’s C-terminal region
with the membrane. This result demonstrates that the lipid interaction of the C-terminal site is
driven by the peptide’s hydrophobic potential and not by the acidic head groups of the
Discussion
72
phospholipids. It is assumed that in vivo the C-terminal arm is exposed to the cytosolic
environment and available even in the closed protein conformation. Therefore, the C-terminal
arm may tether vinculin to the cell membrane and/ or facilitate its interaction with other focal
adhesion components such as talin. Site-directed mutagenesis revealed that intermolecular
interactions between the vinculin-tail and lipids are controlled by surface exposed, basic
residues [82]. Furthermore, a change of two residues (RK 1060/61 Q) located in the C-
terminal lipid-binding site significantly impaired the interaction of the vinculin-tail with
vesicles containing acidic phospholipids. DSC measurements using a 21-residue peptide
carrying this mutation indicated, however, a higher lipid insertion potential for DMPC/
DMPG vesicles (see chapter 3.1.2). Thus, changing residues from positively charged (R, K) to
uncharged (Q) increased the hydrophobic character of this peptide and resulted in increased
interactions with moderately charged lipid vesicles. The in comparison to the unmutated C-
terminal arm increased hydrophobicity may also explain why the lipid binding of vinculin-tail
(RK 1060/61 Q) was suppressed [82]. A conformational change in vinculin-tail triggered by
the mutations may bury the C-terminal arm inside the helical bundle, which results in the loss
of the lipid anchor as membrane interaction site and reduces lipid binding.
4.2 Vinculin´s lipid anchor is involved in beta-sheet formation The protein crystals revealed no well defined secondary structure for the C-terminal arm
which hosts the lipid anchor [55; 78-79]. Protein crystals are often obtained under non-
physiological pH and the proteins are compressed due to crystal packing with minimal solvent
(water) molecules around their surface. Furthermore, a protein crystal displays only one
possible protein conformation which gives no information about flexibility and
conformational changes of the peptide during their function. In order to determine a
secondary structure of the C-terminal arm, the local conformational flexibility of this part was
explored using molecular dynamics simulations (see Chapter 3.2). The observations indicated
for nominally “basic” and “neutral” conditions an anti-parallel beta-sheet followed by an
unstructured C-terminus. In contrast, for an “acidic” environment no stable secondary
structural elements emerged and the C-terminus remained unstructured. This result indicates
that the secondary structure of the C-terminal arm depends on the ratio of positively to
negatively charged residues. As seen in repeated simulations, changing the ionic conditions
after each 10ns, the anti-parallel beta-sheet secondary structure was not influenced by its
previous starting conformation. However, the MD-study exploring the conformational
behavior of the full length vinculin-tail domain under nominal “neutral” settings,
Discussion
73
demonstrated that no anti-parallel beta-sheet emerged at the C-terminal arm (refer to chapter
3.2.2). The vinculin-tail simulation shows that the C-terminus is hindered more sterically by
the rest of protein. This implies that a conformational shift of the vinculin-tail, induced by a
local pH-shift or other interaction partners such as paxillin and actin, is necessary to free the
C-terminal arm before beta-sheet formation. Experimental support for the secondary structure
formation of the 21 residue peptide comes from CD spectroscopy (refer to chapter 3.3). An
alignment using the DICHROWEB database, indicates also two beta-strands and unstructured
parts of the C-terminal region suggesting that at a pH of 7.5 an beta-sheet conformation is
preferred. Together with the conformations observed from the MD-simulations, the following
summary of the C-terminal region secondary structure can be made: (i) residues 2–6 form an
initial beta-strand that represents the first part (residues 1045–1051) and (ii) residues 7–12
complete the beta-strand of the C-terminus of the tail domain (residues 1052–1057). The CD-
spectrum measured in presence of lipid vesicles revealed a totally different absorption curve
in comparison to the measurement without the lipid vesicles. However, the database analysis
obtained also an anti-parallel beta-sheet and mostly unstructured parts for the peptide
representing vinculin´s last 21 residues. This suggests a conformational shift of the peptide,
representing vinculin´s C-terminal arm, during lipid membrane interaction. The difference in
spectral characteristics of the 21-residue peptide during CD spectroscopy in the presence/
absence of lipid vesicles can be explained by the high variance of beta-sheet structures. This
could be due to parallel and anti-parallel orientations and different twists which are reflected
in different backbone angles [89; 120]. In order to determine the orientation of the C-terminal
peptide during lipid membrane interaction, solid state NMR-measurements were performed
(see chapter 3.4). The 15N label was placed in leucine at position 11, which is part of the beta
strand. The result may indicate membrane insertion for the beta-sheet secondary structure
element. However, to verify those results, further NMR measurements are needed with
radioactive labeled residues on different positions. A different solvent for the peptide is also
necessary. Because of the poor solubility in polar solvent at high concentrations, vinculin´s C-
terminal peptide was dissolved in formic acid. However, formic acid has the ability to
denature and degrade peptides. So it is not known if the peptide still exists and has the same
secondary structure, as determined under the conditions of CD-spectroscopy measurement. In
conclusion, this study indicates that the C-terminal 21 residues of the vinculin-tail have the
capability to associate with, or indeed insert into, artificial lipid membranes probably by
forming an anti-parallel beta-sheet.
Discussion
74
4.3 Vinculin´s lipid anchor regulates cell stiffness Mechanical tension between the interconnected assemblies of the extracellular matrix and the
cytoskeleton play a critical role in determining cell structure and function. It was
demonstrated that forces generated by and in a cell regulate many biological functions. This
implies the involvement of key molecules which recognize and generate tension for specific
cellular features [13]. Changes in cell tension influence the molecular structure and
biochemical activity of the cell. To determine the effect of the lipid anchor in respect to cell
mechanics, several EGFP-linked vinculin and vinculin-tail constructs, deficient in lipid
binding, were designed and transiently transfected in MEF-vin(-/-) cells. As demonstrated in
chapter 3.5, the N-terminal EGFP label has no obvious effect on FA localization of full length
vinculin (residues 1-1066) and vinculinΔC (residues 1-1052) molecules in MEF cells. In
contrast, MEF-vin(-/-) cells transfected with the EGFP-vtail (residues 858-1066) and -vtailΔC
(residues 858-1052) constructs showed no well-defined localization to the focal adhesions
(see chapter 3.5.2). Besides focal adhesion localization, the used EGFP-vtail and EGFP–
vtailΔC constructs also covered the actin stress-fibers outside of the focal adhesions. Because
of the lack of the vinculin-head domain, all the binding sites localized at the tail region,
including the actin binding sites, were freely accessible. This exposure of the sites could
explain the interaction of the vinculin-tail constructs with the actin filaments. In the inactive
or closed conformation, the vinculin-head masks the cryptic actin-binding sites and prevents
actin binding [55]. Only in its open or activated form, vinculin is available at the focal
adhesions [52]. This localization ensures that vinculin is connected with actin filaments only
in the focal adhesions. The covered actin filaments in both MEF-vtail and MEF-vtailΔC cells
may explain the magnetic tweezer results. The determination of the binding strength between
ECM-coated surfaces and cells displayed a higher bead detachment for MEF-vtail and MEF-
vtailΔC in comparison to MEF-vin(-/-) (see chapter 3.6.2). The vtail or vtailΔC which were
localized on the actin-filaments outside the focal adhesions may impair the actin turnover.
This disturbs the integrin clustering and FA formation around the bead. However, the stiffness
of MEF-vtail and MEF-vtailΔC was increased by ~20 % in comparison to MEF-vin(-/-) cells.
The stiffness calculation, excluding the detached beads, obtained the same stiffness values for
the MEF-vtail and MEF-vtailΔC cells as for MEF-wt cells, suggesting that the vinculin-tail is
responsible for vinculin´s mechano-coupling properties. This is consistent with results
obtained from magnetic tweezer using F9-vinculin(-/-) cells, stably expressing the vinculin-
tail domain. It was reported that the F9-vtail cells restore the wildtype phenotype
mechanically [65-66]. In contrast to MEF cells, expressing vinculin (residues 1-1066) and
Discussion
75
vinculinΔC (residues 1-1052) (see chapter 3.6.1), no difference could be detected between
MEF-vtail and MEF-vtailΔC, suggesting vinculin´s lipid anchor is involved in the regulation
of vinculin´s accessibility for other binding partners. The so-called MEF-resc cells showed
almost identical mechanical properties to their wildtype counterparts. However, MEF-vin(-/-)
cells transiently transfected with EGFP-vinculinΔC were less stiff compared to MEF-resc
cells. It is known that the vinculin-tail lacking the lipid anchor showed impaired interaction
with acidic phospholipid vesicles [78; 81]. This indicates that the difference in stiffness of
MEF-vinΔC compared to MEF-resc and MEF-wt cell might be due to the linkage of the cell
membrane with the lipid anchor of vinculin. Support for this assumption also comes from the
DSC measurements (see chapter 3.1.1) [80]. The results of the binding strength measurements
are also indicative for an involvement of the lipid anchor in controlling cell mechanical
properties. With increased pulling force, more beads detached from vinculinΔC cells than
from MEF-wt or rescue cells during the measurements. At high forces, they displayed almost
the same level of disrupted fibronectin-coated beads as the vinculin knock-out cells,
suggesting that the lipid anchor is important for the binding strength to ECM-coated surfaces.
Furthermore, the reduced binding strength of MEF-vinΔC cells to the beads may also explain
the reduced stiffness in MEF-vinΔC cells. As demonstrated in figure 4-1, the relative stiffness
values of MEF-vinΔC cells, determined at 1 nN force, were significantly reduced in
comparison to MEF-wt and MEF-resc cells. In contrast, the values of MEF-vinΔC cells
obtained at 10nN were as stiff as MEF-wt and MEF-resc cells. For the stiffness calculation at
1nN all measured beads were used, including those which could not sustain high forces and
detach during the measurement. In general, such weakly bound beads show increased
displacements during tweezers measurements in comparison to tightly bound beads [69; 65].
After these weakly bound beads had been removed from the analysis, as demonstrated for the
measurement at 10 nN force, MEF-vinΔC cells were indistinguishable from MEF-wt and
MEF-resc cells in terms of average stiffness (see chapter 3.6.1). This indicates that vinculin
must be anchored to the cell membrane with its C-terminal arm to guarantee a well
established connection between the bead and the FA, suggesting that MEF-vinΔC cells are
less stiff because of the reduced binding strength to ECM-coated surfaces. This connection
between binding strength and stiffness leads to the conclusion that there are two
subpopulations of beads in MEF-vinΔC cells. The well attached and the weakly bound ones.
The well attached beads show the same stiffness than the beads on MEF-wt and MEF-resc
cells. This suggests a difference in the turnover rate for vinculin and vinculinΔC molecules at
the focal adhesions. However, FRAP measurements in transfected MEF-vin(-/-) cells revealed
Discussion
76
no significant differences between the turnover rate of EGFP vinculin and vinculinΔC (refer
to chapter 3.9), indicating that the lipid anchor influences the recruitment of other FA proteins
which in turn control the turnover rate of the whole focal adhesion. Indeed, it was
demonstrated that the FA turnover rate in cells expressing vinculinΔC was decreased in
comparison to cells expressing intact vinculin [81]. However, FRAP measurements performed
in Keratoncytes revealed that the vinculin turnover rate in nascent FA is higher than in mature
adhesions (In: Modulation of vinculin exchange dynamics regulates adhesion site maturation
and adhesion strength; by Christoph Möhl, Norbert Kirchgeßner, Claudia Schäfer, Kevin
Küpper, Gerold Diez, Wolfgang H. Goldmann, Rudolf Merkel, Bernd Hoffmann; submitted to
Cell Motility and Cytoskeleton). During the FRAP-measurements, performed in MEF-cells
(see chapter 3.9), it could not be distinguished between nascent and mature focal adhesions.
This might also explain the similar turnover rate for vinculin and vinculinΔC, measured in
MEF-vin(-/-) cells.
0
0,2
0,4
0,6
0,8
1
1,2
1,4
11 nN 10 nN
Rel
ativ
e st
iffne
ss
MEF-wtMEF-rescMEF-vin∆CMEF-vin(-/-)
Figure 4-1: Relative stiffness of the different MEF-cells at 1nN and 10nN, respectively. The stiffness calculation at 1nN force includes also beads that detached during the measurement, whilst at 10nN only the beads could be used which remained attached. The error bars represent the standard error (SEM).
4.4 The lipid anchor affects traction generation via phosphotyrosine 1065 Additional evidence that the lipid anchor is important for regulating the mechanical properties
in cells came from 2D-traction microscopy measurements. The strain energy for MEF-wt and
rescue cells was about 8-fold higher compared to MEF-vin(-/-). In contrast, MEF-vinΔC cells
generate only 2 times more strain energy in comparison to MEF-vin(-/-) cells (refer to chapter
3.10.1). This indicates that the lipid anchor is important for internal force generation which in
turn controls FA growth and maintenance [109]. Cell adhesive forces are believed to be due to
Discussion
77
the actin ECM connection and the myosin-II driven force development [5; 7; 15]. It was
demonstrated when blocking the contractility with substances such as ML-7, BDM, KT5926
or by the over-expression of peptides like caldesmon (inhibits actin-dependent myosin-II
ATPases activity), that it leads to dissolution of focal contacts [27; 109]. This could explain
the decreased number of focal adhesions in MEF-vinΔC in comparison to MEF-wt and rescue
cells (refer to chapter 3.8).
Chandrasekar and co-workers demonstrated that the introduction of point mutations in Helix 3
(residues 935 – 978) and the C-terminal arm (residues 1045-1066) of the vinculin-tail result
also in impaired lipid membrane interaction [82]. To test the influence of each lipid binding
site to force generation, additional 2D-traction experiments were performed using MEF-vin(-
/-) cells, transfected with vinculin-H3 (mutated residues: K952, K956, R963, R966 to Q),
vinculin-CT (mutated residues: RK1060/61 to Q) and vinculin-LD (vinculin-H3 + CT). MEF-
LD as well as MEF–CT cells developed ~ 50 % less tractions in comparison to MEF-resc
cells. This was in the same range as MEF-vinΔC cells. However, MEF-H3 cells displayed
almost the same strain energy than MEF-resc cells (see chapter 3.10.2). It suggests that only
the membrane association of the lipid anchor (residue 1052-1066) is important for force
generation, whilst the lipid binding of vinculin-H3 has no influence on cell mechanical
behavior. Recently, it was demonstrated, that src phosphorylates vinculin on tyrosine 100 and
1065 [71]. It was reported when preventing c-src phosphorylation, the interaction of vinculin
with the Arp2/3 subunit p34Arc is inhibited [72]. This results in impaired cell spreading and
migration [71-72]. One of those src phosphorylation sites, tyrosine 1065, is localized in the
lipid anchor. To test the influence of Y1065 due to traction generation, 2D-traction
microscopy experiments of MEF cells transfected with vinculinY1065F were performed.
Indeed, MEF-Y1065 cells generated the same reduced strain energy compared to MEF-resc.
Those reduced values of vinculinY1065F cells were comparable to MEF-LD, MEF-CT and
MEF-vinΔC cells (refer to chapter 3.10.2). In vitro experiments demonstrated that acidic
phospholipid vesicles stimulate the phosphorylation of vinculin by src kinase [73-74]. In the
presence of divalent cations and acidic phospholipids such as phosphatidyl-inositol (PI) or
phosphatidyl-glycerol (PG), the phosphorylated fraction of vinculin was elevated more than
10-fold in comparison to vinculin in the absence of lipid vesicles. However, the
phosphorylation levels of other src targets such as alpha-actinin and casein were not affected
by artificial membranes, which implies a conformational shift of vinculin prior to
phosphorylation [73]. Indeed, protease digestion assays of vinculin in the presence and
absence of such lipid vesicles demonstrated a higher degradation of the membrane bound
Discussion
78
vinculin peptide in comparison to unbound vinculin, suggesting that the interaction of
vinculin with the lipid membrane induces a conformational shift [74]. This implies that the
membrane interaction of the lipid anchor provides vinculin´s src-dependent phosphorylation
which in turn is necessary for traction generation.
4.5 The Model Previous studies indicate that the loss of the lipid anchor as well as the introduction of six
surface-exposed basic residues on the vinculin-tail result in impaired membrane interaction
[78; 81-82]. Cells transfected with vinculin variants containing those alterations displayed
reduced cell migration and adhesion, because of their decreased FA turnover rate [81-82]. The
function of vinculin´s lipid binding is still a controversy. Chandrasekar and co-workers
demonstrated a link between the interaction of vinculin with acidic phospholipids and focal
adhesion disassembly [82]. An increase of PIP2, caused by an over-expression of PIP 5-kinase
alpha, resulted in vinculin delocalization from the focal adhesion whilst vinculin deficient in
lipid binding remained in the focal adhesion complex. In contrast, it was proposed that
membrane binding triggers the displacement of vinculin´s head domain from the tail which
could lead to its activation and FA recruitment [59-60; 74]. There is also evidence for other
proteins to activate vinculin. It is conceivable that there are several intermediate
conformations of vinculin, dependent on its binding partners. In vitro and in vivo experiments
demonstrated that talin, alpha-actinin as well as actin alter also vinculin´s conformation and
activate the molecule [52-54]. Indeed, no difference in FA localization between vinculin and
vinculinΔC cells could be detected. The shape and size of adhesions were also similar. FRAP
measurements in transfected MEF-vin(-/-) cells revealed also no significant differences
between the turnover rate of EGFP vinculin and vinculinΔC (refer to chapter 3.9), indicating
the lipid anchor influences the recruitment of other FA components which in turn controls the
turnover rate of the entire adhesion plaque.
Furthermore, an involvement in cell mechanical regulation was only determined for the C-
terminal lipid anchor. In contrast, the lipid binding site localized on helix 3 (residues 935 –
978) had no influence on cell mechanical behavior. This may imply that each lipid binding
site of vinculin has its own purpose and function, independent from each other. Crystal
structure analysis of vinculin indicates that the lipid anchor region is not buried inside the
vinculin molecule in the closed conformation [55; 79]. This allows the lipid anchor to interact
with the hydrophobic core of lipid membranes even in the closed or auto-inhibited state of
vinculin (current work) [80]. In summary, it is possible that the lipid anchor of vinculin
Discussion
79
guides the molecule to the cell membrane where it becomes prepared for FA recruitment.
Localized at the membrane, vinculin is in close proximity to other focal components like talin
and actin, which have also the ability to activate vinculin for focal adhesion recruitment. The
impaired membrane localization of vinculin lacking the lipid anchor may result in an
incomplete conformational switch that could result in an impaired connection to the actin
cytoskeleton in MEF-vinΔC cells.
The following model was drafted from the described results. As demonstrated in figure 4-2,
the vinculin molecule becomes activated by talin, actin or phospholipids. Further, the
anchorage of vinculin´s C-terminal to the cell membrane prepares the molecule for src-
dependent phosphorylation [73-74]. Identified c-src targets of vinculin are tyrosine 100 and
tyrosine 1065 [71]. The last tyrosine is part of the lipid anchor. The phosphorylation of
vinculin by c-src enables the molecule to bind Arp2/3, an actin nucleator and crosslinker [72;
121]. This interaction may reinforce the connection of the focal adhesion complex to the actin
cytoskeleton, which enables the cells to generate the appropriate pre-stress and strain energy
for adequate cell adhesion and migration. Therefore the lipid anchor regulates the mechano-
coupler vinculin via membrane binding dependent src-phosphorylation, followed by Arp2/3
recruitment. Thus, additional experiments are necessary to prove this mechanism.
Figure 4-2: Lipid binding of vinculin via its lipid anchor provides the molecule for src-dependent phosphorylation.
4.6 Outlook The lipid anchor regulates cell adhesion and migration by src dependent phosphorylation. The
MEF-cells are a model system for further investigations. In the next step, the phosphorylation
Discussion
80
level of tyrosine 1065 should be determined by Western blotting for all those mutant variants
used in this study. Since it is known that tyrosine 100 is also phosphorylated by src, it is also
necessary to characterize its influence on cell mechanical behavior, using the magnetic
tweezer and 2D-traction microscopy. Additional FRAP measurement with vinculin and
vinculin deficient in lipid binding as well as deficient in tyrosine 100/ 1065 phosphorylation
in nascent focal adhesions should reveal a difference in the turnover rate. It is also necessary
to determine the influence of phosphorylation to membrane binding. Molecular dynamics
simulations of the phosphorylated and unphosphorylated vinculin-tail together with lipid
membranes are another possibility to obtain further informations. Additional pull-down
experiments with the phosphorylated vinculin-tail, incubated with lipid vesicles, would also
be helpful to clarify these questions.
References
81
5 References [1] J.A. Alberts B., Lewis J., Raff M., Roberts K., Walter P., The Cell 4th edition, Garland
Science, New York, 2004. [2] D.E. Ingber, Cellular mechanotransduction: putting all the pieces together again. Faseb J
20 (2006) 811-27. [3] D. Stamenovic, Micromechanical foundations of pulmonary elasticity. Physiol Rev 70
(1990) 1117-34. [4] D.J. Tschumperlin, G. Dai, I.V. Maly, T. Kikuchi, L.H. Laiho, A.K. McVittie, K.J. Haley,
C.M. Lilly, P.T. So, D.A. Lauffenburger, R.D. Kamm, and J.M. Drazen, Mechanotransduction through growth-factor shedding into the extracellular space. Nature 429 (2004) 83-6.
[5] A.D. Bershadsky, N.Q. Balaban, and B. Geiger, Adhesion-dependent cell mechanosensitivity. Annu Rev Cell Dev Biol 19 (2003) 677-95.
[6] A. Bershadsky, M. Kozlov, and B. Geiger, Adhesion-mediated mechanosensitivity: a time to experiment, and a time to theorize. Curr Opin Cell Biol 18 (2006) 472-81.
[7] B. Geiger, and A. Bershadsky, Exploring the neighborhood: adhesion-coupled cell mechanosensors. Cell 110 (2002) 139-42.
[8] E. Zamir, and B. Geiger, Molecular complexity and dynamics of cell-matrix adhesions. J Cell Sci 114 (2001) 3583-90.
[9] S. Pokutta, and W.I. Weis, The cytoplasmic face of cell contact sites. Curr Opin Struct Biol 12 (2002) 255-62.
[10] A.S. Yap, M.S. Crampton, and J. Hardin, Making and breaking contacts: the cellular biology of cadherin regulation. Curr Opin Cell Biol 19 (2007) 508-14.
[11] D.L. Stokes, Desmosomes from a structural perspective. Curr Opin Cell Biol 19 (2007) 565-71.
[12] B.S. Hahn, and M. Labouesse, Tissue integrity: hemidesmosomes and resistance to stress. Curr Biol 11 (2001) R858-61.
[13] N. Wang, J.P. Butler, and D.E. Ingber, Mechanotransduction across the cell surface and through the cytoskeleton. Science 260 (1993) 1124-7.
[14] C.S. Izzard, and L.R. Lochner, Cell-to-substrate contacts in living fibroblasts: an interference reflexion study with an evaluation of the technique. J Cell Sci 21 (1976) 129-59.
[15] A.D. Bershadsky, C. Ballestrem, L. Carramusa, Y. Zilberman, B. Gilquin, S. Khochbin, A.Y. Alexandrova, A.B. Verkhovsky, T. Shemesh, and M.M. Kozlov, Assembly and mechanosensory function of focal adhesions: experiments and models. Eur J Cell Biol 85 (2006) 165-73.
[16] W.H. Goldmann, Kinetic determination of focal adhesion protein formation. Biochem Biophys Res Commun 271 (2000) 553-7.
[17] D.A. Calderwood, Y. Fujioka, J.M. de Pereda, B. Garcia-Alvarez, T. Nakamoto, B. Margolis, C.J. McGlade, R.C. Liddington, and M.H. Ginsberg, Integrin beta cytoplasmic domain interactions with phosphotyrosine-binding domains: a structural prototype for diversity in integrin signaling. Proc Natl Acad Sci U S A 100 (2003) 2272-7.
[18] B. Geiger, and A. Bershadsky, Assembly and mechanosensory function of focal contacts. Curr Opin Cell Biol 13 (2001) 584-92.
[19] D.R. Critchley, Focal adhesions - the cytoskeletal connection. Curr Opin Cell Biol 12 (2000) 133-9.
[20] W.H. Ziegler, R.C. Liddington, and D.R. Critchley, The structure and regulation of vinculin. Trends Cell Biol 16 (2006) 453-60.
[21] C.E. Turner, Paxillin and focal adhesion signalling. Nat Cell Biol 2 (2000) E231-6.
References
82
[22] B. Geiger, A. Bershadsky, R. Pankov, and K.M. Yamada, Transmembrane crosstalk between the extracellular matrix--cytoskeleton crosstalk. Nat Rev Mol Cell Biol 2 (2001) 793-805.
[23] H.B. Wang, M. Dembo, S.K. Hanks, and Y. Wang, Focal adhesion kinase is involved in mechanosensing during fibroblast migration. Proc Natl Acad Sci U S A 98 (2001) 11295-300.
[24] G. von Wichert, B. Haimovich, G.S. Feng, and M.P. Sheetz, Force-dependent integrin-cytoskeleton linkage formation requires downregulation of focal complex dynamics by Shp2. Embo J 22 (2003) 5023-35.
[25] N.Q. Balaban, U.S. Schwarz, D. Riveline, P. Goichberg, G. Tzur, I. Sabanay, D. Mahalu, S. Safran, A. Bershadsky, L. Addadi, and B. Geiger, Force and focal adhesion assembly: a close relationship studied using elastic micropatterned substrates. Nat Cell Biol 3 (2001) 466-72.
[26] M. Chrzanowska-Wodnicka, and K. Burridge, Rho-stimulated contractility drives the formation of stress fibers and focal adhesions. J Cell Biol 133 (1996) 1403-15.
[27] D.M. Helfman, E.T. Levy, C. Berthier, M. Shtutman, D. Riveline, I. Grosheva, A. Lachish-Zalait, M. Elbaum, and A.D. Bershadsky, Caldesmon inhibits nonmuscle cell contractility and interferes with the formation of focal adhesions. Mol Biol Cell 10 (1999) 3097-112.
[28] I. Kaverina, O. Krylyshkina, K. Beningo, K. Anderson, Y.L. Wang, and J.V. Small, Tensile stress stimulates microtubule outgrowth in living cells. J Cell Sci 115 (2002) 2283-91.
[29] D. Riveline, E. Zamir, N.Q. Balaban, U.S. Schwarz, T. Ishizaki, S. Narumiya, Z. Kam, B. Geiger, and A.D. Bershadsky, Focal contacts as mechanosensors: externally applied local mechanical force induces growth of focal contacts by an mDia1-dependent and ROCK-independent mechanism. J Cell Biol 153 (2001) 1175-86.
[30] L.P. Cramer, Molecular mechanism of actin-dependent retrograde flow in lamellipodia of motile cells. Front Biosci 2 (1997) d260-70.
[31] D. Choquet, D.P. Felsenfeld, and M.P. Sheetz, Extracellular matrix rigidity causes strengthening of integrin- cytoskeleton linkages. Cell 88 (1997) 39-48.
[32] C.G. Galbraith, K.M. Yamada, and M.P. Sheetz, The relationship between force and focal complex development. J Cell Biol 159 (2002) 695-705.
[33] R.J. Pelham, Jr., and Y. Wang, Cell locomotion and focal adhesions are regulated by substrate flexibility. Proc Natl Acad Sci U S A 94 (1997) 13661-5.
[34] C.M. Lo, H.B. Wang, M. Dembo, and Y.L. Wang, Cell movement is guided by the rigidity of the substrate. Biophys J 79 (2000) 144-52.
[35] C.D. Nobes, and A. Hall, Rho, rac, and cdc42 GTPases regulate the assembly of multimolecular focal complexes associated with actin stress fibers, lamellipodia, and filopodia. Cell 81 (1995) 53-62.
[36] K. Rottner, A. Hall, and J.V. Small, Interplay between Rac and Rho in the control of substrate contact dynamics. Curr Biol 9 (1999) 640-8.
[37] M.A. del Pozo, L.S. Price, N.B. Alderson, X.D. Ren, and M.A. Schwartz, Adhesion to the extracellular matrix regulates the coupling of the small GTPase Rac to its effector PAK. EMBO J 19 (2000) 2008-14.
[38] D.L. Scott, G. Diez, and W.H. Goldmann, Protein-lipid interactions: correlation of a predictive algorithm for lipid-binding sites with three-dimensional structural data. Theoretical Biology and Medical Modelling 3 (2006).
[39] L.M. Machesky, R.D. Mullins, H.N. Higgs, D.A. Kaiser, L. Blanchoin, R.C. May, M.E. Hall, and T.D. Pollard, Scar, a WASp-related protein, activates nucleation of actin filaments by the Arp2/3 complex. Proc Natl Acad Sci U S A 96 (1999) 3739-44.
References
83
[40] H. Miki, H. Yamaguchi, S. Suetsugu, and T. Takenawa, IRSp53 is an essential intermediate between Rac and WAVE in the regulation of membrane ruffling. Nature 408 (2000) 732-5.
[41] D.C. Edwards, L.C. Sanders, G.M. Bokoch, and G.N. Gill, Activation of LIM-kinase by Pak1 couples Rac/Cdc42 GTPase signalling to actin cytoskeletal dynamics. Nat Cell Biol 1 (1999) 253-9.
[42] C.E. Turner, M.C. Brown, J.A. Perrotta, M.C. Riedy, S.N. Nikolopoulos, A.R. McDonald, S. Bagrodia, S. Thomas, and P.S. Leventhal, Paxillin LD4 motif binds PAK and PIX through a novel 95-kD ankyrin repeat, ARF-GAP protein: A role in cytoskeletal remodeling. J Cell Biol 145 (1999) 851-63.
[43] C. Ballestrem, B. Hinz, B.A. Imhof, and B. Wehrle-Haller, Marching at the front and dragging behind: differential alphaVbeta3-integrin turnover regulates focal adhesion behavior. J Cell Biol 155 (2001) 1319-32.
[44] D. Tsuruta, M. Gonzales, S.B. Hopkinson, C. Otey, S. Khuon, R.D. Goldman, and J.C. Jones, Microfilament-dependent movement of the beta3 integrin subunit within focal contacts of endothelial cells. FASEB J 16 (2002) 866-8.
[45] R. Pankov, E. Cukierman, B.Z. Katz, K. Matsumoto, D.C. Lin, S. Lin, C. Hahn, and K.M. Yamada, Integrin dynamics and matrix assembly: tensin-dependent translocation of alpha(5)beta(1) integrins promotes early fibronectin fibrillogenesis. J Cell Biol 148 (2000) 1075-90.
[46] E. Zamir, M. Katz, Y. Posen, N. Erez, K.M. Yamada, B.Z. Katz, S. Lin, D.C. Lin, A. Bershadsky, Z. Kam, and B. Geiger, Dynamics and segregation of cell-matrix adhesions in cultured fibroblasts. Nat Cell Biol 2 (2000) 191-6.
[47] K. Kimura, M. Ito, M. Amano, K. Chihara, Y. Fukata, M. Nakafuku, B. Yamamori, J. Feng, T. Nakano, K. Okawa, A. Iwamatsu, and K. Kaibuchi, Regulation of myosin phosphatase by Rho and Rho-associated kinase (Rho-kinase). Science 273 (1996) 245-8.
[48] M. Amano, Y. Fukata, and K. Kaibuchi, Regulation and functions of Rho-associated kinase. Exp Cell Res 261 (2000) 44-51.
[49] T. Ishizaki, Y. Morishima, M. Okamoto, T. Furuyashiki, T. Kato, and S. Narumiya, Coordination of microtubules and the actin cytoskeleton by the Rho effector mDia1. Nat Cell Biol 3 (2001) 8-14.
[50] S. Munevar, Y.L. Wang, and M. Dembo, Regulation of mechanical interactions between fibroblasts and the substratum by stretch-activated Ca2+ entry. J Cell Sci 117 (2004) 85-92.
[51] N.D. Gallant, K.E. Michael, and A.J. Garcia, Cell adhesion strengthening: contributions of adhesive area, integrin binding, and focal adhesion assembly. Mol Biol Cell 16 (2005) 4329-40.
[52] H. Chen, D.M. Cohen, D.M. Choudhury, N. Kioka, and S.W. Craig, Spatial distribution and functional significance of activated vinculin in living cells. J Cell Biol 169 (2005) 459-70.
[53] D.M. Cohen, B. Kutscher, H. Chen, D.B. Murphy, and S.W. Craig, A conformational switch in vinculin drives formation and dynamics of a talin-vinculin complex at focal adhesions. J Biol Chem 281 (2006) 16006-15.
[54] H. Chen, D.M. Choudhury, and S.W. Craig, Coincidence of actin filaments and talin is required to activate vinculin. J Biol Chem 281 (2006) 40389-98.
[55] C. Bakolitsa, D.M. Cohen, L.A. Bankston, A.A. Bobkov, G.W. Cadwell, L. Jennings, D.R. Critchley, S.W. Craig, and R.C. Liddington, Structural basis for vinculin activation at sites of cell adhesion. Nature 430 (2004) 583-6.
[56] T. Izard, G. Evans, R.A. Borgon, C.L. Rush, G. Bricogne, and P.R. Bois, Vinculin activation by talin through helical bundle conversion. Nature 427 (2004) 171-5.
References
84
[57] P.R. Bois, B.P. O'Hara, D. Nietlispach, J. Kirkpatrick, and T. Izard, The vinculin binding sites of talin and alpha-actinin are sufficient to activate vinculin. J Biol Chem 281 (2006) 7228-36.
[58] P.R. Bois, R.A. Borgon, C. Vonrhein, and T. Izard, Structural dynamics of alpha-actinin-vinculin interactions. Mol Cell Biol 25 (2005) 6112-22.
[59] A.P. Gilmore, and K. Burridge, Regulation of vinculin binding to talin and actin by phosphatidyl-inositol-4-5-bisphosphate. Nature 381 (1996) 531-5.
[60] R.P. Johnson, and S.W. Craig, The carboxy-terminal tail domain of vinculin contains a cryptic binding site for acidic phospholipids. Biochem Biophys Res Commun 210 (1995) 159-64.
[61] P.A. Steimle, J.D. Hoffert, N.B. Adey, and S.W. Craig, Polyphosphoinositides inhibit the interaction of vinculin with actin filaments. J Biol Chem 274 (1999) 18414-20.
[62] J.L. Coll, A. Ben-Ze'ev, R.M. Ezzell, J.L. Rodriguez Fernandez, H. Baribault, R.G. Oshima, and E.D. Adamson, Targeted disruption of vinculin genes in F9 and embryonic stem cells changes cell morphology, adhesion, and locomotion. Proc Natl Acad Sci U S A 92 (1995) 9161-5.
[63] R.M. Ezzell, W.H. Goldmann, N. Wang, N. Parasharama, and D.E. Ingber, Vinculin promotes cell spreading by mechanically coupling integrins to the cytoskeleton. Exp Cell Res 231 (1997) 14-26.
[64] W. Xu, H. Baribault, and E.D. Adamson, Vinculin knockout results in heart and brain defects during embryonic development. Development 125 (1998) 327-37.
[65] C.T. Mierke, P. Kollmannsberger, D. Paranhos-Zitterbart, J. Smith, B. Fabry, and W.H. Goldmann, Mechano-coupling and regulation of contractility by the vinculin tail domain. . Biophys J 94 (2008) 661-70.
[66] W. Xu, J.L. Coll, and E.D. Adamson, Rescue of the mutant phenotype by reexpression of full-length vinculin in null F9 cells; effects on cell locomotion by domain deleted vinculin. J Cell Sci 111 ( Pt 11) (1998) 1535-44.
[67] W.H. Goldmann, R. Galneder, M. Ludwig, W. Xu, E.D. Adamson, N. Wang, and R.M. Ezzell, Differences in elasticity of vinculin-deficient F9 cells measured by magnetometry and atomic force microscopy. Exp Cell Res 239 (1998) 235-42.
[68] W.H. Goldmann, R. Galneder, M. Ludwig, A. Kromm, and R.M. Ezzell, Differences in F9 and 5.51 cell elasticity determined by cell poking and atomic force microscopy. FEBS Lett 424 (1998) 139-42.
[69] F.J. Alenghat, B. Fabry, K.Y. Tsai, W.H. Goldmann, and D.E. Ingber, Analysis of cell mechanics in single vinculin-deficient cells using a magnetic tweezer. Biochem Biophys Res Commun 277 (2000) 93-9.
[70] M.C. Subauste, O. Pertz, E.D. Adamson, C.E. Turner, S. Junger, and K.M. Hahn, Vinculin modulation of paxillin-FAK interactions regulates ERK to control survival and motility. J Cell Biol 165 (2004) 371-81.
[71] Z. Zhang, G. Izaguirre, S.Y. Lin, H.Y. Lee, E. Schaefer, and B. Haimovich, The phosphorylation of vinculin on tyrosine residues 100 and 1065, mediated by SRC kinases, affects cell spreading. Mol Biol Cell 15 (2004) 4234-47.
[72] S. Moese, M. Selbach, V. Brinkmann, A. Karlas, B. Haimovich, S. Backert, and T.F. Meyer, The Helicobacter pylori CagA protein disrupts matrix adhesion of gastric epithelial cells by dephosphorylation of vinculin. Cell Microbiol 9 (2007) 1148-61.
[73] S. Ito, N. Richert, and I. Pastan, Phospholipids stimulate phosphorylation of vinculin by the tyrosine-specific protein kinase of Rous sarcoma virus. Proc Natl Acad Sci U S A 79 (1982) 4628-31.
[74] S. Ito, D.K. Werth, N.D. Richert, and I. Pastan, Vinculin phosphorylation by the src kinase. Interaction of vinculin with phospholipid vesicles. J Biol Chem 258 (1983) 14626-31.
References
85
[75] J. Smith, G. Diez, A.H. Klemm, V. Schewkunow, and W.H. Goldmann, CapZ-lipid membrane interactions: a computer analysis. Theor Biol Med Model 3 (2006) 30.
[76] M. Tempel, W.H. Goldmann, G. Isenberg, and E. Sackmann, Interaction of the 47-kDa talin fragment and the 32-kDa vinculin fragment with acidic phospholipids: a computer analysis. Biophys J 69 (1995) 228-41.
[77] R.P. Johnson, V. Niggli, P. Durrer, and S.W. Craig, A conserved motif in the tail domain of vinculin mediates association with and insertion into acidic phospholipid bilayers. Biochemistry 37 (1998) 10211-22.
[78] C. Bakolitsa, J.M. de Pereda, C.R. Bagshaw, D.R. Critchley, and R.C. Liddington, Crystal structure of the vinculin tail suggests a pathway for activation. Cell 99 (1999) 603-13.
[79] R.A. Borgon, C. Vonrhein, G. Bricogne, P.R. Bois, and T. Izard, Crystal structure of human vinculin. Structure (Camb) 12 (2004) 1189-97.
[80] G. Diez, F. List, J. Smith, W.H. Ziegler, and W.H. Goldmann, Direct evidence of vinculin tail–lipid membrane interaction in beta-sheet conformation. Biochem Biophys Res Commun 373 (2008) 69-73.
[81] R.M. Saunders, M.R. Holt, L. Jennings, D.H. Sutton, I.L. Barsukov, A. Bobkov, R.C. Liddington, E.A. Adamson, G.A. Dunn, and D.R. Critchley, Role of vinculin in regulating focal adhesion turnover Eur J Cell Biol 85 (2006) 487-500.
[82] I. Chandrasekar, T.E. Stradal, M.R. Holt, F. Entschladen, B.M. Jockusch, and W.H. Ziegler, Vinculin acts as a sensor in lipid regulation of adhesion-site turnover. J Cell Sci 118 (2005) 1461-72.
[83] D.R. Joe Sambrook, Molecular Cloning, Cold spring harbour Laboratory, Cold spring harbour, 2001.
[84] D. Hanahan, Studies on transformation of Escherichia coli with plasmids. J Mol Biol 166 (1983) 557-80.
[85] W. Wang, and B.A. Malcolm, Two-stage PCR protocol allowing introduction of multiple mutations, deletions and insertions using QuikChange Site-Directed Mutagenesis. Biotechniques 26 (1999) 680-2.
[86] F. Sanger, S. Nicklen, and A.R. Coulson, DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci U S A 74 (1977) 5463-7.
[87] R.B. Merrifield, Solid Phase Peptide Synthesis. I. The Synthesis of a Tetrapeptide. J. Am. Chem. Soc. 85 (1963) 6.
[88] W.H. Goldmann, B. Bechinger, and T.P. Lele, Cytoskeletal proteins at the lipid membrane. in: T.H. T., and O.-L. A., (Eds.), Planar lipid bilayers (BLM's) and their applications, Elsevier, 2007, pp. 227-255.
[89] L. Whitmore, and B.A. Wallace, DICHROWEB, an online server for protein secondary structure analyses from circular dichroism spectroscopic data. Nucleic Acids Res 32 (2004) W668-73.
[90] A. Lobley, L. Whitmore, and B.A. Wallace, DICHROWEB: an interactive website for the analysis of protein secondary structure from circular dichroism spectra. Bioinformatics 18 (2002) 211-2.
[91] M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72 (1976) 248-54.
[92] U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227 (1970) 680-5.
[93] A.R. Leach, Molecular Modelling: Principles and Applications, Pearson Education EMA, 2001.
[94] D. Van Der Spoel, E. Lindahl, B. Hess, G. Groenhof, A.E. Mark, and H.J. Berendsen, GROMACS: fast, flexible, and free. J Comput Chem 26 (2005) 1701-18.
References
86
[95] E. Lindahl, B. Hess, and D. Van Der Spoel, GROMACS 3.0: a package for molecular simulation and trajectory analysis. J Mol Model 7 (2001) 306-17.
[96] H.J.C. Berendsen, J.P.M. Postma, W.F. van Gunsteren, and J. Hermans, Interaction models for water in relation to protein hydration. in: B. Pullman, (Ed.), Intermolecular Forces, Reidel, Dordrecht, 1981.
[97] C. Oostenbrink, T.A. Soares, N.F. van der Vegt, and W.F. van Gunsteren, Validation of the 53A6 GROMOS force field. Eur Biophys J 34 (2005) 273-84.
[98] B. Hess, H. Bekker, H.J.C. Berendsen, and J.G.E.M. Fraaije, LINCS: A Linear Constraint Solver for Molecular Simulations. J Comput Chem 18 (1997) 1463-72.
[99] H.J.C. Berendsen, J.P.M. Postma, W.F. Van Gunsteren, A. DiNola, and J.R. Haak, Molecular dynamics with coupling to an external bath J Chem Phys 81 (1984) 3684-90.
[100] W. Kabsch, and C. Sander, Dictionary of protein secondary structure: pattern recognition of hydrogen-bonded and geometrical features. Biopolymers 22 (1983) 2577-637.
[101] J.H. Ward, Hierarchical Grouping for Evaluating Clustering Methods. J Am Statistical Association 58 (1963) 236-44.
[102] R. Mojena, Hierarchical grouping methods and stopping rules: An evaluation. Computer Journal 20 (1977) 359-63.
[103] P. Kollmannsberger, and B. Fabry, High-Force Magnetic Tweezers with Force Feedback for Biological Applications. Rev Sci Instrum 78 (2007) 114301-1-6.
[104] J.P. Butler, I.M. Tolic-Norrelykke, B. Fabry, and J.J. Fredberg, Traction fields, moments, and strain energy that cells exert on their surroundings. Am J Physiol Cell Physiol 282 (2002) C595-605.
[105] C. Metzner, C. Raupach, D. Paranhos Zitterbart, and B. Fabry, Simple model of cytoskeletal fluctuations. Phys Rev E 76 (2007) 021925-1-12.
[106] M. Tempel, W.H. Goldmann, C. Dietrich, V. Niggli, T. Weber, E. Sackmann, and G. Isenberg, Insertion of filamin into lipid membranes examined by calorimetry, the film balance technique, and lipid photolabeling. Biochemistry 33 (1994) 12565-72.
[107] V. Schewkunow, K.P. Sharma, G. Diez, A.H. Klemm, P.C. Sharma, and W.H. Goldmann, Thermodynamic evidence of non-muscle myosin II-lipid-membrane interaction. Biochem Biophys Res Commun 366 (2008) 500-5.
[108] J.D. Humphries, P. Wang, C. Streuli, B. Geiger, M.J. Humphries, and C. Ballestrem, Vinculin controls focal adhesion formation by direct interactions with talin and actin. J Cell Biol 179 (2007) 1043-57.
[109] N.Q. Balaban, U.S. Schwarz, D. Riveline, P. Goichberg, G. Tzur, I. Sabanay, D. Mahalu, S. Safran, A. Bershadsky, L. Addadi, and B. Geiger, Force and focal adhesion assembly: a close relationship studied using elastic micropatterned substrates. Nat Cell Biol 3 (2001) 466-72.
[110] M. Dembo, and Y.L. Wang, Stresses at the cell-to-substrate interface during locomotion of fibroblasts. Biophys J 76 (1999) 2307-16.
[111] M. Hyvonen, M.J. Macias, M. Nilges, H. Oschkinat, M. Saraste, and M. Wilmanns, Structure of the binding site for inositol phosphates in a PH domain. EMBO J 14 (1995) 4676-85.
[112] V. Niggli, Structural properties of lipid-binding sites in cytoskeletal proteins. Trends Biochem Sci 26 (2001) 604-11.
[113] J.H. Hurley, and S. Misra, Signaling and subcellular targeting by membrane-binding domains. Annu Rev Biophys Biomol Struct 29 (2000) 49-79.
[114] D. Marsh, and T. Pali, The protein-lipid interface: perspectives from magnetic resonance and crystal structures. Biochim Biophys Acta 1666 (2004) 118-41.
References
87
[115] V. Niggli, D.P. Dimitrov, J. Brunner, and M.M. Burger, Interaction of the cytoskeletal component vinculin with bilayer structures analyzed with a photoactivatable phospholipid. J Biol Chem 261 (1986) 6912-8.
[116] V. Niggli, S. Kaufmann, W.H. Goldmann, T. Weber, and G. Isenberg, Identification of functional domains in the cytoskeletal protein talin. Eur J Biochem 224 (1994) 951-7.
[117] W.H. Goldmann, J.L. Niles, and M.A. Arnaout, Interaction of purified human proteinase 3 (PR3) with reconstituted lipid bilayers. Eur J Biochem 261 (1999) 155-62.
[118] W.H. Goldmann, J.M. Teodoridis, C.P. Sharma, J.L. Alonso, and G. Isenberg, Fragments from alpha-actinin insert into reconstituted lipid bilayers. Biochem Biophys Res Commun 264 (1999) 225-9.
[119] V. Niggli, L. Sommer, J. Brunner, and M.M. Burger, Interaction in situ of the cytoskeletal protein vinculin with bilayers studied by introducing a photoactivatable fatty acid into living chicken embryo fibroblasts. Eur J Biochem 187 (1990) 111-7.
[120] L. Whitmore, and B.A. Wallace, Protein secondary structure analyses from circular dichroism spectroscopy: Methods and reference databases. Biopolymers 89 (2008) 392-400.
[121] K.A. DeMali, C.A. Barlow, and K. Burridge, Recruitment of the Arp2/3 complex to vinculin: coupling membrane protrusion to matrix adhesion. J Cell Biol 159 (2002) 881-91.
Appendix
88
6 Appendix
6.1 Visualization of focal contacts using the TEM The focal adhesion sites connect the actin cytoskeleton via the integrin receptor to the
extracellular matrix. To investigate the density of focal adhesion plaques, we characterized
the FA distribution of paxillin proteins in mouse embryonic fibroblasts. The cells were
cultured and prepared for TEM characterization as described under 2.2.2.7. The gold labeled
paxillin molecules are near a longitudinal structure of 200-400 nm length and 10-20nm
diameter, which could represent actin fibers. Judged by the density of the beads around the
actin fiber, it is believed to observe a focal contact. The next step clearly is to extend the
investigations to MEF-vin(-/-) cells, for visualizing any differences between MEF-wt and
knock out cells to obtain differences in the density of paxillin at the focal adhesions.
Figure 6-1: Electronmicroscopic images of MEF-wt cells at 50.000x (A) and 130.000x (B) magnification. B represents the magnified spot of image A, labeled with the red circle. The arrows in picture B mark the gold labeled secondary antibodies which represent the paxillin molecules. Note, that the accumulation of the paxillin molecules may represent a focal adhesion complex.
Appendix
89
6.2 pcEGFP-N2 expression vector
MCS: NheI 851 GCTTATCGAA ATTAATACGA CTCACTATAG GGAGACCCAA GCTGGCTAGC 901 GCTACCGGTC GCCACCATGG TGAGCAAGGG CGAGGAGCTG TTCACCGGGG M V S K G E E L F T G V Frame 2 951 TGGTGCCCAT CCTGGTCGAG CTGGACGGCG ACGTAAACGG CCACAAGTTC V P I L V E L D G D V N G H K F Frame 2 1001 AGCGTGTCCG GCGAGGGCGA GGGCGATGCC ACCTACGGCA AGCTGACCCT S V S G E G E G D A T Y G K L T L Frame 2 1051 GAAGTTCATC TGCACCACCG GCAAGCTGCC CGTGCCCTGG CCCACCCTCG K F I C T T G K L P V P W P T L V Frame 2 1101 TGACCACCCT GACCTACGGC GTGCAGTGCT TCAGCCGCTA CCCCGACCAC T T L T Y G V Q C F S R Y P D H Frame 2 1151 ATGAAGCAGC ACGACTTCTT CAAGTCCGCC ATGCCCGAAG GCTACGTCCA M K Q H D F F K S A M P E G Y V Q Frame 2 1201 GGAGCGCACC ATCTTCTTCA AGGACGACGG CAACTACAAG ACCCGCGCCG E R T I F F K D D G N Y K T R A E Frame 2 1251 AGGTGAAGTT CGAGGGCGAC ACCCTGGTGA ACCGCATCGA GCTGAAGGGC V K F E G D T L V N R I E L K G Frame 2
Appendix
90
1301 ATCGACTTCA AGGAGGACGG CAACATCCTG GGGCACAAGC TGGAGTACAA I D F K E D G N I L G H K L E Y N Frame 2 1351 CTACAACAGC CACAACGTCT ATATCATGGC CGACAAGCAG AAGAACGGCA Y N S H N V Y I M A D K Q K N G I Frame 2 1401 TCAAGGTGAA CTTCAAGATC CGCCACAACA TCGAGGACGG CAGCGTGCAG K V N F K I R H N I E D G S V Q Frame 2 1451 CTCGCCGACC ACTACCAGCA GAACACCCCC ATCGGCGACG GCCCCGTGCT L A D H Y Q Q N T P I G D G P V L Frame 2 1501 GCTGCCCGAC AACCACTACC TGAGCACCCA GTCCGCCCTG AGCAAAGACC L P D N H Y L S T Q S A L S K D P Frame 2 1551 CCAACGAGAA GCGCGATCAC ATGGTCCTGC TGGAGTTCGT GACCGCCGCC N E K R D H M V L L E F V T A A Frame 2 BsrGI AflII XhoI 1601 GGGATCACTC TCGGCATGGA CGAGCTGTAC AAGTCCGGAC TCCTTAAGCT G I T L G M D E L Y K S G L L K L Frame 2 XbaI ApaI PmeI 1651 CGAGTCTAGA GGGCCCGTTT AAACCCGCTG ATCAGCCTCG ACTGTGCCTT E S R G P V * Frame 2
Appendix
91
6.3 cDNA of vinculin mouse BsiWI MluI PvuI AflIII BsiEI PfoI BsrFI 1 ATGCCAGTGT TTCATACGCG TACGATCGAG AGCATCCTGG AGCCGGTGGC TACGGTCACA AAGTATGCGC ATGCTAGCTC TCGTAGGACC TCGGCCACCG +H3N- M P V F H T R T I E S I L E P V A Frame 1 1% 6 % % % SexAI DraIII BssSI 51 GCAGCAGATC TCGCACCTGG TGATTATGCA CGAGGAGGGC GAGGTGGACG CGTCGTCTAG AGCGTGGACC ACTAATACGT GCTCCTCCCG CTCCACCTGC Q Q I S H L V I M H E E G E V D G Frame 1 ? 20 a ? a a % % % BsgI EagI BglI BsiEI 101 GCAAAGCCAT TCCTGACCTC ACCGCGCCCG TAGCCGCCGT GCAGGCGGCC CGTTTCGGTA AGGACTGGAG TGGCGCGGGC ATCGGCGGCA CGTCCGCCGG K A I P D L T A P V A A V Q A A Frame 1 40 NciI 151 GTCAGCAACC TCGTCCGGGT TGGAAAAGAG ACTGTTCAGA CCACTGAGGA CAGTCGTTGG AGCAGGCCCA ACCTTTTCTC TGACAAGTCT GGTGACTCCT V S N L V R V G K E T V Q T T E D Frame 1 60 201 TCAGATTCTG AAGAGAGATA TGCCACCAGC CTTTATTAAG GTTGAAAATG AGTCTAAGAC TTCTCTCTAT ACGGTGGTCG GAAATAATTC CAACTTTTAC Q I L K R D M P P A F I K V E N A Frame 1 80 HindIII 251 CTTGCACCAA GCTTGTTCAG GCAGCCCAGA TGCTTCAGTC AGACCCATAC GAACGTGGTT CGAACAAGTC CGTCGGGTCT ACGAAGTCAG TCTGGGTATG C T K L V Q A A Q M L Q S D P Y Frame 1 100 301 TCGGTTCCTG CGCGGGATTA CCTCATTGAC GGCTCTAGGG GAATCCTTTC AGCCAAGGAC GCGCCCTAAT GGAGTAACTG CCGAGATCCC CTTAGGAAAG S V P A R D Y L I D G S R G I L S Frame 1 V-head D1 (6-252) % % BbvCI 351 TGGCACATCT GACCTACTGC TTACCTTCGA TGAGGCTGAG GTTCGTAAAA ACCGTGTAGA CTGGATGACG AATGGAAGCT ACTCCGACTC CAAGCATTTT G T S D L L L T F D E A E V R K I Frame 1 120 % ApoI 401 TTATTAGGGT TTGCAAAGGA ATTTTGGAAT ATCTTACAGT GGCAGAGGTA AATAATCCCA AACGTTTCCT TAAAACCTTA TAGAATGTCA CCGTCTCCAT I R V C K G I L E Y L T V A E V Frame 1 140
Appendix
92
Tth111I 451 GTGGAAACTA TGGAAGACTT GGTCACTTAC ACAAAGAATC TTGGGCCAGG CACCTTTGAT ACCTTCTGAA CCAGTGAATG TGTTTCTTAG AACCCGGTCC V E T M E D L V T Y T K N L G P G Frame 1 160 HincII 501 AATGACTAAG ATGGCCAAAA TGATTGATGA GAGACAGCAG GAGTTGACTC TTACTGATTC TACCGGTTTT ACTAACTACT CTCTGTCGTC CTCAACTGAG M T K M A K M I D E R Q Q E L T H Frame 1 180 % % ? For1 DraIII FalI 551 ACCAGGAACA CCGTGTGATG TTGGTGAACT CAATGAACAC TGTCAAAGAG TGGTCCTTGT GGCACACTAC AACCACTTGA GTTACTTGTG ACAGTTTCTC Q E H R V M L V N S M N T V K E Frame 1 ? 200 601 CTGCTTCCTG TTCTCATTTC AGCTATGAAG ATTTTTGTTA CAACCAAAAA GACGAAGGAC AAGAGTAAAG TCGATACTTC TAAAAACAAT GTTGGTTTTT L L P V L I S A M K I F V T T K N Frame 1 HindIII ApoI 651 CTCAAAAAAC CAAGGAATAG AAGAAGCTTT GAAAAATCGA AATTTTACTG GAGTTTTTTG GTTCCTTATC TTCTTCGAAA CTTTTTAGCT TTAAAATGAC S K N Q G I E E A L K N R N F T V Frame 1 220 Rev1 701 TAGAAAAGAT GAGTGCTGAA ATTAACGAGA TCATTCGTGT GTTACAACTC ATCTTTTCTA CTCACGACTT TAATTGCTCT AGTAAGCACA CAATGTTGAG E K M S A E I N E I I R V L Q L Frame 1 240 A 751 ACTTCCTGGG ATGAAGATGC CTGGGCCAGC AAGGACACTG AAGCCATGAA TGAAGGACCC TACTTCTACG GACCCGGTCG TTCCTGTGAC TTCGGTACTT T S W D E D A W A S K D T E A M K Frame 1 ? a 260 HaeII HgaI AfeI BsaHI PflMI 801 GAGAGCGCTG GCGTCCATAG ACTCCAAATT GAACCAGGCC AAAGGTTGGC CTCTCGCGAC CGCAGGTATC TGAGGTTTAA CTTGGTCCGG TTTCCAACCG R A L A S I D S K L N Q A K G W L Frame 1 280 TspGWI PasI 851 TCCGTGACCC CAATGCCTCC CCAGGGGATG CTGGAGAGCA GGCCATCAGG AGGCACTGGG GTTACGGAGG GGTCCCCTAC GACCTCTCGT CCGGTAGTCC R D P N A S P G D A G E Q A I R Frame 1 300 BsgI 901 CAGATCTTAG ATGAAGCTGG AAAAGTTGGT GAACTTTGTG CAGGCAAGGA GTCTAGAATC TACTTCGACC TTTTCAACCA CTTGAAACAC GTCCGTTCCT Q I L D E A G K V G E L C A G K E Frame 1 AvrII 951 ACGCAGGGAG ATCCTAGGAA CCTGCAAAAT GCTAGGGCAG ATGACTGACC TGCGTCCCTC TAGGATCCTT GGACGTTTTA CGATCCCGTC TACTGACTGG R R E I L G T C K M L G Q M T D Q Frame 1 320
Appendix
93
BmrI 1001 AAGTGGCTGA CCTCCGAGCC AGAGGACAAG GAGCTTCCCC AGTGGCCATG TTCACCGACT GGAGGCTCGG TCTCCTGTTC CTCGAAGGGG TCACCGGTAC V A D L R A R G Q G A S P V A M Frame 1 340
TaiI HpyCH4IV BanII BmgBI 1051 CAGAAGGCCC AGCAAGTGTC TCAGGGGCTC GACGTGCTTA CCGCCAAAGT GTCTTCCGGG TCGTTCACAG AGTCCCCGAG CTGCACGAAT GGCGGTTTCA Q K A Q Q V S Q G L D V L T A K V Frame 1 360 BsmI AlfI 1101 GGAGAATGCA GCTCGGAAGC TGGAAGCCAT GACGAACTCA AAGCAGAGCA CCTCTTACGT CGAGCCTTCG ACCTTCGGTA CTGCTTGAGT TTCGTCTCGT E N A A R K L E A M T N S K Q S I Frame 1 380 BamHI EciI PflMI 1151 TTGCAAAGAA GATTGATGCT GCCCAGAATT GGCTGGCGGA TCCAAATGGT AACGTTTCTT CTAACTACGA CGGGTCTTAA CCGACCGCCT AGGTTTACCA A K K I D A A Q N W L A D P N G Frame 1 V-head D2 (253-485) 400 1201 GGACCTGAGG GAGAAGAACA GATTCGAGGG GCTTTGGCTG AAGCTCGGAA CCTGGACTCC CTCTTCTTGT CTAAGCTCCC CGAAACCGAC TTCGAGCCTT G P E G E E Q I R G A L A E A R K Frame 1 BsrBI 1251 GATTGCAGAA TTATGTGATG ATCCTAAGGA GAGAGATGAC ATCCTCCGCT CTAACGTCTT AATACACTAC TAGGATTCCT CTCTCTACTG TAGGAGGCGA I A E L C D D P K E R D D I L R S Frame 1 420 1301 CCCTTGGAGA GATAGCTGCT CTGACCTCTA AACTAGGAGA CTTGCGAAGA GGGAACCTCT CTATCGACGA GACTGGAGAT TTGATCCTCT GAACGCTTCT L G E I A A L T S K L G D L R R Frame 1 440 XhoI PspXI 1351 CAGGGGAAAG GAGACTCGCC AGAGGCTCGA GCCTTGGCTA AACAAGTGGC GTCCCCTTTC CTCTGAGCGG TCTCCGAGCT CGGAACCGAT TTGTTCACCG Q G K G D S P E A R A L A K Q V A Frame 1 460 PstI 1401 GACGGCACTA CAGAACCTGC AGACCAAAAC CAACAGGGCC GTGGCCAACA CTGCCGTGAT GTCTTGGACG TCTGGTTTTG GTTGTCCCGG CACCGGTTGT T A L Q N L Q T K T N R A V A N S Frame 1 % % 480
Appendix
94
XhoI PvuII PspXI 1451 GCAGACCTGC CAAAGCAGCT GTCCACCTCG AGGGCAAGAT TGAACAGGCG CGTCTGGACG GTTTCGTCGA CAGGTGGAGC TCCCGTTCTA ACTTGTCCGC R P A K A A V H L E G K I E Q A Frame 1 V % % K 500 DrdI TaqII 1501 CAGCGGTGGA TTGATAACCC CACAGTAGAT GACCGTGGAG TCGGTCAGGC GTCGCCACCT AACTATTGGG GTGTCATCTA CTGGCACCTC AGCCAGTCCG Q R W I D N P T V D D R G V G Q A Frame 1 % % TspGWI BbvCI BspHI 1551 TGCCATCCGT GGACTTGTGG CTGAGGGGCA TCGGCTGGCC AATGTCATGA ACGGTAGGCA CCTGAACACC GACTCCCCGT AGCCGACCGG TTACAGTACT A I R G L V A E G H R L A N V M M Frame 1 520 PpuMI BsmFI BslFI BslFI FalI AccI 1601 TGGGACCTTA TCGCCAAGAT CTTCTTGCCA AATGTGACCG TGTAGACCAG ACCCTGGAAT AGCGGTTCTA GAAGAACGGT TTACACTGGC ACATCTGGTC G P Y R Q D L L A K C D R V D Q Frame 1 540 V-head D3 (493-717) PvuII BlpI 1651 CTAACAGCTC AGCTGGCTGA CCTGGCTGCC CGAGGGGAGG GGGAGAGTCC GATTGTCGAG TCGACCGACT GGACCGACGG GCTCCCCTCC CCCTCTCAGG L T A Q L A D L A A R G E G E S P Frame 1 560 1701 TCAGGCGAGA GCACTTGCAT CCCAGCTTCA GGACTCCTTA AAGGATCTTA AGTCCGCTCT CGTGAACGTA GGGTCGAAGT CCTGAGGAAT TTCCTAGAAT Q A R A L A S Q L Q D S L K D L K Frame 1 580 BpuEI BciVI 1751 AAGCCCAGAT GCAGGAAGCT ATGACTCAAG AGGTATCCGA TGTTTTCAGC TTCGGGTCTA CGTCCTTCGA TACTGAGTTC TCCATAGGCT ACAAAAGTCG A Q M Q E A M T Q E V S D V F S Frame 1 600 BsaXI BsaXI BsaXI for2 BsaXI 1801 GATACTACAA CTCCTATCAA GCTGTTGGCA GTAGCCGCCA CTGCTCCTCC CTATGATGTT GAGGATAGTT CGACAACCGT CATCGGCGGT GACGAGGAGG D T T T P I K L L A V A A T A P P Frame 1 1851 TGATGCACCC AATAGGGAAG AGGTATTTGA TGAAAGGGCA GCCAATTTTG ACTACGTGGG TTATCCCTTC TCCATAAACT ACTTTCCCGT CGGTTAAAAC D A P N R E E V F D E R A A N F E Frame 1 620 S 1901 AAAACCATTC AGGAAGGCTT GGAGCCACAG CAGAGAAGGC GGCTGCTGTT TTTTGGTAAG TCCTTCCGAA CCTCGGTGTC GTCTCTTCCG CCGACGACAA N H S G R L G A T A E K A A A V Frame 1 640
Appendix
95
Rev2 BsmI 1951 GGTACTGCCA ATAAATCGAC AGTGGAAGGC ATTCAGGCAT CTGTGAAGAC CCATGACGGT TATTTAGCTG TCACCTTCCG TAAGTCCGTA GACACTTCTG G T A N K S T V E G I Q A S V K T Frame 1 ? a % % a 660 2001 AGCCCGAGAA CTCACTCCCC AGGTCATCTC CGCTGCTCGG ATCTTACTGA TCGGGCTCTT GAGTGAGGGG TCCAGTAGAG GCGACGAGCC TAGAATGACT A R E L T P Q V I S A A R I L L R Frame 1 680 BstEII BsaI 2051 GGAACCCTGG TAACCAGGCT GCTTATGAAC ATTTTGAGAC CATGAAGAAC CCTTGGGACC ATTGGTCCGA CGAATACTTG TAAAACTCTG GTACTTCTTG N P G N Q A A Y E H F E T M K N Frame 1 700 2101 CAGTGGATTG ATAATGTTGA AAAAATGACA GGGCTGGTGG ACGAGGCTAT GTCACCTAAC TATTACAACT TTTTTACTGT CCCGACCACC TGCTCCGATA Q W I D N V E K M T G L V D E A I Frame 1 % 2151 TGATACCAAG TCTCTGTTGG ATGCTTCTGA AGAAGCAATT AAAAAAGACC ACTATGGTTC AGAGACAACC TACGAAGACT TCTTCGTTAA TTTTTTCTGG D T K S L L D A S E E A I K K D L Frame 1 ? 720 NcoI SspI 2201 TGGACAAGTG TAAGGTAGCC ATGGCCAATA TTCAGCCTCA GATGCTGGTC ACCTGTTCAC ATTCCATCGG TACCGGTTAT AAGTCGGAGT CTACGACCAG D K C K V A M A N I Q P Q M L V Frame 1 740 BsaWI 2251 GCTGGAGCAA CCAGTATTGC TCGTCGGGCC AACCGGATTC TGCTGGTTGC CGACCTCGTT GGTCATAACG AGCAGCCCGG TTGGCCTAAG ACGACCAACG A G A T S I A R R A N R I L L V A Frame 1 V-head D4 (719-835) 760 AloI AloI PpuMI 2301 TAAGAGGGAG GTAGAGAACT CTGAGGACCC GAAGTTCCGA GAGGCTGTGA ATTCTCCCTC CATCTCTTGA GACTCCTGGG CTTCAAGGCT CTCCGACACT K R E V E N S E D P K F R E A V K Frame 1 % % % a % % % 780 NcoI 2351 AAGCTGCCTC TGATGAACTG AGCAAAACAA TCTCCCCCAT GGTGATGGAT TTCGACGGAG ACTACTTGAC TCGTTTTGTT AGAGGGGGTA CCACTACCTA A A S D E L S K T I S P M V M D Frame 1 K 800 PfoI 2401 GCCAAGGCTG TGGCTGGAAA CATCTCTGAC CCTGGCCTGC AAAAGAGCTT CGGTTCCGAC ACCGACCTTT GTAGAGACTG GGACCGGACG TTTTCTCGAA A K A V A G N I S D P G L Q K S F Frame 1 810 D EcoRV BamHI 2451 CCTGGACTCA GGATATCGGA TCCTGGGAGC TGTGGCCAAG GTCAGAGAAG GGACCTGAGT CCTATAGCCT AGGACCCTCG ACACCGGTTC CAGTCTCTTC L D S G Y R I L G A V A K V R E A Frame 1 820 Q For3 2501 CCTTCCAACC TCAGGAGCCT GACTTCCCGC CTCCTCCACC AGACCTTGAA GGAAGGTTGG AGTCCTCGGA CTGAAGGGCG GAGGAGGTGG TCTGGAACTT F Q P Q E P D F P P P P P D L E Frame 1
Appendix
96
% 840% % SacI EcoICRI 2551 CAGCTACGAC TAACTGATGA GCTCGCTCCT CCTAAGCCAC CTCTGCCTGA GTCGATGCTG ATTGACTACT CGAGCGAGGA GGATTCGGTG GAGACGGACT Q L R L T D E L A P P K P P L P E Frame 1 860 BsmFI BslFI AloI BslFI AloI 2601 GGGTGAAGTC CCTCCACCCA GGCCCCCACC ACCAGAAGAG AAGGATGAAG CCCACTTCAG GGAGGTGGGT CCGGGGGTGG TGGTCTTCTC TTCCTACTTC G E V P P P R P P P P E E K D E E Frame 1 879 880 2651 AGTTCCCTGA GCAGAAAGCT GGTGAGGTGA TTAACCAGCC AATGATGATG TCAAGGGACT CGTCTTTCGA CCACTCCACT AATTGGTCGG TTACTACTAC F P E Q K A G E V I N Q P M M M Frame 1 896 900 2701 GCCGCCAGGC AGCTCCACGA TGAAGCTCGG AAATGGTCTA GCAAGGGCAA CGGCGGTCCG TCGAGGTGCT ACTTCGAGCC TTTACCAGAT CGTTCCCGTT A A R Q L H D E A R K W S S K G N Frame 1 % ? Vt Helix I ? 913 AlfI 2751 TGACATCATT GCAGCAGCCA AGCGCATGGC TCTGCTGATG GCAGAGATGT ACTGTAGTAA CGTCGTCGGT TCGCGTACCG AGACGACTAC CGTCTCTACA D I I A A A K R M A L L M A E M S Frame 1 920 Vt Helix II KpnI BanI Acc65I 2801 CTCGGCTGGT AAGAGGGGGC AGTGGTACCA AGCGGGCACT TATTCAGTGT GAGCCGACCA TTCTCCCCCG TCACCATGGT TCGCCCGTGA ATAAGTCACA R L V R G G S G T K R A L I Q C Frame 1 P 938 940 %944 % % EcoRV StuI 2851 GCCAAGGATA TCGCCAAGGC CTCTGATGAG GTGACGAGGT TGGCCAAGGA CGGTTCCTAT AGCGGTTCCG GAGACTACTC CACTGCTCCA ACCGGTTCCT A K D I A K A S D E V T R L A K E Frame 1 Vt Helix III 960 2901 GGTTGCCAAG CAGTGCACAG ATAAGCGGAT TAGAACCAAT CTCTTACAGG CCAACGGTTC GTCACGTGTC TATTCGCCTA ATCTTGGTTA GAGAATGTCC V A K Q C T D K R I R T N L L Q V Frame 1 972 % % %975% % 980% % % % 2951 TATGCGAGCG AATCCCAACT ATAAGCACCC AGCTCAAAAT CCTATCCACA ATACGCTCGC TTAGGGTTGA TATTCGTGGG TCGAGTTTTA GGATAGGTGT C E R I P T I S T Q L K I L S T Frame 1 a % % a a % 1000 3001 GTGAAGGCCA CTATGCTGGG CCGGACCAAC ATCAGTGATG AGGAGTCTGA CACTTCCGGT GATACGACCC GGCCTGGTTG TAGTCACTAC TCCTCAGACT V K A T M L G R T N I S D E E S E Frame 1 Vt Helix IV 1005 % 1013%% % % 3051 GCAGGCCACA GAGATGCTGG TTCATAATGC CCAGAACCTC ATGCAGTCTG CGTCCGGTGT CTCTACGACC AAGTATTACG GGTCTTGGAG TACGTCAGAC Q A T E M L V H N A Q N L M Q S V Frame 1 1020 % ? ? Vt Helix V
Appendix
97
3101 TGAAGGAGAC TGTGCGAGAG GCTGAAGCTG CTTCAATCAA AATCCGAACA ACTTCCTCTG ACACGCTCTC CGACTTCGAC GAAGTTAGTT TTAGGCTTGT K E T V R E A E A A S I K I R T Frame 1 % 1040 T 1046% % KpnI BanI Acc65I 3151 GATGCTGGCT TTACTCTGCG CTGGGTCAGA AAGACTCCCT GGTACCAGTA CTACGACCGA AATGAGACGC GACCCAGTCT TTCTGAGGGA CCATGGTCAT D A G F T L R W V R K T P W Y Q * Frame 1 1052 Flexible C-tail 1060 1066 3201 G C
D1: residues 1-252 D2: residues 253-485 D1-D4 Vinculin-head D3: residues 493-717 D4: residues 719-835 Poly-Proline: residues 838-878 D5: residue 898-1066 → also called vinculin-tail
Appendix
98
6.4 List of Figures Front-page: AFM-image of a MEF-cell (kindly provided by Tilman Schäffer) Figure 1-1: Simplified cartoon of a cross section through part of the intestine ......................... 1 Figure 1-2: Cellular mechano responsiveness ........................................................................... 2 Figure 1-3: Schematic of cell contacts visualized in epithelial cells.......................................... 5 Figure 1-4: Illustration of focal adhesions. ................................................................................ 6 Figure 1-5: Schematics of the rac/ rho dependent focal adhesion formation............................. 9 Figure 1-6: Schematic of the vinculin molecule ...................................................................... 11 Figure 1-7: The vinculin-tail with its three potential lipid binding sites.................................. 13 Figure 2-1: Cloning strategy for the generation of the pEGFP expression system.................. 21 Figure 2-2: Cloning strategy for the vinculin cDNA. .............................................................. 22 Figure 2-3: Sample preparation for TEM measuremnts........................................................... 26 Figure 2-4: Phase transition thermogram of DMPG/DMPC vesicles.. .................................... 28 Figure 2-5 A schematic of the DSC apparatus. ........................................................................ 29 Figure 2-6: Simulated solid-state NMR spectra....................................................................... 31 Figure 3-1: Vinculin peptide representing vincuin´s C-terminal arm. ..................................... 39 Figure 3-2: Interaction of the mutated vinculin-tail peptide (RK1060/61 Q) with MLVs. ..... 40 Figure 3-3: Secondary structure of vinculin´s C-terminal arm under “basic” conditions........ 43 Figure 3-4: Secondary structure of vinculin´s C-terminal arm under “neutral” conditions..... 44 Figure 3-5: Secondary structure of vinculin´s the C-terminal arm under “acidic” conditions 45 Figure 3-6: MD-simulations of the 21 residue peptide ............................................................ 46 Figure 3-7: Secondary structure of the vinculin-tail (residue 890-1066)................................. 47 Figure 3-8: Results from CD-spectroscopic measurements. .................................................... 48 Figure 3-9: Result of the NMR measurement. ........................................................................ 50 Figure 3-10: Expression of vinculin and vinculinΔC in MEF cells. ........................................ 51 Figure 3-11: Expression of EGFP linked vinculin-tail and vinculin-tailΔC. ........................... 52 Figure 3-12: Expression of vinculin and vinculinY1065F in MEF-vin(-/-) cells. ................... 53 Figure 3-13: Co-localization of the auto-fluorescent beads with the focal adhesions. ............ 54 Figure 3-14: Result of magnetic tweezer measurements ......................................................... 56 Figure 3-15: Differential cell stiffness of transfected and non-transfected MEF cells ............ 57 Figure 3-16: Geometric mean of the power-law exponent b .................................................. 58 Figure 3-17: Magnetic tweezer measurements of MEF-vtail and MEF-vtailΔC..................... 60 Figure 3-18: Differential cell stiffness of MEF-vtail and vtailΔC cells................................... 61 Figure 3-19: The mean of the power-law exponent b for MEF-vtail and MEF-vtailΔC. ........ 62 Figure 3-20: Spreading area of transfected and non-tranfected MEF cells.............................. 63 Figure 3-21: Determination of focal adhesions (FA) per cell line ........................................... 64 Figure 3-22: Turn over rate of EGFP-vinculin and vinculinΔC .............................................. 65 Figure 3-23: Results of the 2D-traction microscopy measurements ........................................ 67 Figure 3-24: Strain energy measurements of MEF-LD, MEF-CT, MEF-H3, MEF-Y1065F.. 68 Figure 4-1: Relative stiffness of the different MEF-cells at 1 nN and 10 nN .......................... 76 Figure 4-2: Model..................................................................................................................... 79 Figure 6-1: Electronmicroscopic images of MEF-wt cells. ..................................................... 88
Acknowledgements
99
Acknowledgements I extend my sincere thanks to Prof. Dr. Wolfgang H. Goldmann for giving me this great
opportunity, to write my PhD thesis under his guidance. He was always very encouraging and
supportive to my work.
I am grateful to Prof. Dr. Hojatollah Vali for his wonderful supervision, encouragement, and
support during my term at McGill University. I enjoyed working in his group and I am sure
this experience will help me in my future career.
I am thankful for Prof. Dr. Ben Fabry´s financial support and valued discussions throughout
this work. My special thanks go to Dr. James Smith, for his supervision and guidance.
I am grateful to my dear colleague Anna H. Klemm. Without her help and value advices this
work would not have been possible.
I greatly acknowledge the help from Philip Kollmannsberger, Thorsten Koch, Daniel
Paranhos Zitterbart, Carina Raupach and Stefan Münster with useful discussions and
suggestions.
Many thanks to Dr. Fereshte Azzari, University of McGill, for her kind help and support to
this work. I would like to include Aleksandra A. Krukiewicz, Barbara Reischel, Ulrike Scholz
and Christine Albert for her good technical support in this regard.
I am thankful to Dr. Wolfgang H. Ziegler, IZKF Leipzig, for the collaboration, help and
support.
I thank Prof. Dr. Hannappel and Prof. Dr. Brandstätter for their acceptance to be involved in
the thesis committee. Thank you to Prof. Dr. Tilman Schäffer for the AFM-image of a MEF-
cell.
My sincere thanks go to Nicole Dietzel for helping me with my thesis. Many thanks also to
her husband Reinhard Dietzel for the technical assistance.
I thank Dr. Bernd Hoffmann and Christoph Möhl, Forschungszentrum Jülich, as well as Prof.
Dr. Burkard Bechinger (University of Strassbourg) for the useful discussions and support. I
also thank Felix List, University of Regensburg and Volker Wirth for their help and good
chats.
My sincere thanks go to all my good friends (Barbara, Bruno, Michi, Kerstin, Martin, Fariba,
Christian, Ralf and Uwe). All of you deserve a PhD degree.
I am grateful to my family for their constant encouragement, support and trust on me which
helped me to get to this stage.
I am very happy to have Martina in my life; I thank her for her unconditional love, patience,
encouragement and support.
Acknowledgements
100
101
Curriculum Vitae Personal Datas: Name: Gerold Diez Adress: Friedrich-Ebert-Str 26
97318 Kitzingen Germany
Date of Birth: 22nd January 1978 Telephone: 049131/8525602 E-mail: [email protected] Education: 09/1984 - 06/1993 Qualifizierter Hauptschulabschluss: Hauptschule Nüdlingen 09/1993 - 06/1995 Staatlich geprüfter Sozialpfleger: Berufsschule Schweinfurt 09/1995 - 06/1996 Realschulabschluss: Berufsaufbauschule Münnerstadt 09/1996 - 06/1999 Abitur: Bayernkolleg Schweinfurt 07/1999 - 07/2000 Wehrersatzdienst 10/2000 - 08/2005 Study of Biochemistry
08/2005 Diploma in Biochemistry: Department of Biochemistry I, University of Regensburg (Germany)
09/2005 - 12/2008 PhD-Student at the Department of Medical Physics and
Technology, Friedrich-Alexander-Universität Erlangen/ Nuremberg (Germany)
12/2008 PhD in Biophysics