VU Research Portal

Elasticity of Biomolecules Baclayon, M.

2014

document version Publisher's PDF, also known as Version of record

Link to publication in VU Research Portal

citation for published version (APA) Baclayon, M. (2014). Elasticity of Biomolecules: probing, pushing and pulling using atomic force microscopy.

General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal ?

Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

E-mail address: [email protected]

Download date: 10. Oct. 2021 Elasticity of Biomolecules

probing, pushing and pulling

using atomic force microscopy

VRIJE UNIVERSITEIT

Elasticity of Biomolecules

probing, pushing and pulling using atomic force microscopy

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad Doctor aan de Vrije Universiteit Amsterdam, op gezag van de rector magniﬁcus prof.dr. F.A. van der Duyn Schouten, in het openbaar te verdedigen ten overstaan van de promotiecommissie van de Faculteit der Exacte Wetenschappen op maandag 20 januari 2014 om 11.45 uur in de aula van de universiteit, De Boelelaan 1105

door

Marian Baclayon Salumbides

geboren te Butuan, Agusan del Norte, Filipijnen promotor: prof.dr.ir. G.J.L. Wuite copromotor: dr. W.H. Roos for my family

The task is, not so much to see what no one has yet seen; but to think what nobody has yet thought, about that which everybody sees.

Erwin Schrödinger† This thesis was reviewed by the members of the reviewing committee:

prof.dr. A.J.R. Heck (Utrecht University, The Netherlands) prof.dr. M. Rief (Technische Universität München, Germany) prof.dr. A. Zlotnick (Indiana University, USA) dr. K. Blank (Radboud University Nijmegen, The Netherlands) dr.ir. S.J.T. van Noort (Leiden University, The Netherlands) dr. J.P. van Ulsen (Vrije Universiteit Amsterdam, The Netherlands)

The work described in this thesis is supported by the STW-administered NanoSci-E+ grant.

The cover is the author’s interpretation of elasticity. †This quote is also attributed to the German philosopher, Arthur Schopenhauer.

1 Introduction 1 1.1 Single-molecule imaging ...... 1 1.2 Single-molecule manipulation and force spectroscopy ...... 2 1.3 Physics of biomolecules ...... 4 1.3.1 Elasticity of biopolymers, proteins and DNA ...... 5 1.3.2 Elasticity of intra/extra-cellular ﬁlamentous networks . . . . . 7 1.3.3 Elasticity of viruses, vesicles and cells ...... 9 1.4 Outline of this thesis ...... 11

2 Sampling protein form and function 15 2.1 Introduction ...... 16 2.2 Principles of atomic force microscopy ...... 17 2.3 High resolution imaging ...... 21 2.4 Indentation experiments ...... 23 2.5 Unfolding experiments ...... 26 2.6 Conclusion ...... 27

3 Imaging at the (supra)molecular scale 29 3.1 Introduction ...... 30 3.2 Methodology ...... 31 3.3 Compaction of mitochondrial DNA by TFAM ...... 35 3.4 Size of nanocolloidal-albumin for tracing tumuors ...... 39 3.5 The pH-regulated assembly of Norwalk virus-like particles ...... 43 3.6 Summary ...... 50

4 Microscopy and force spectroscopy of viruses 51 4.1 Introduction ...... 52 4.2 Atomic Force Microscopy ...... 53 4.3 Imaging of viruses and genome release ...... 54 CONTENTS

4.3.1 AFM imaging of viruses ...... 54 4.3.2 Studying cell-independent genome release by AFM ...... 57 4.4 Imaging virus-cell interactions ...... 60 4.5 Manipulation of viruses in air ...... 62 4.6 Manipulation of viruses in liquid ...... 64 4.6.1 Effect of genome packaging ...... 66 4.6.2 Changes in material properties during maturation ...... 68 4.6.3 Effect of structural modifications in the capsid ...... 68 4.7 Combining experiments and modelling ...... 70 4.8 Conclusion ...... 71

5 Prestress in Norwalk virus nanoparticles 73 5.1 Introduction ...... 74 5.2 Materials and methods ...... 76 5.2.1 Protein preparation ...... 76 5.2.2 Mass spectrometry and native gel electrophoresis ...... 76 5.2.3 Atomic force microscopy ...... 77 5.2.4 Nanoindentation data analysis ...... 77 5.3 Results and discussion ...... 80 5.3.1 Native mass spectra of the particles ...... 80 5.3.2 Size of the Norwalk virus-like particles ...... 80 5.3.3 Elastic properties of the particles ...... 80 5.3.4 Buckling and breakage versus plasticity ...... 83 5.3.5 Comparison to linear elasticity theory ...... 84 5.3.6 Prestress increased rigidity and strength of NVLP ...... 86 5.4 Conclusion ...... 87

6 Folding-driven secretion mechanism of autotransporters 89 6.1 Introduction ...... 90 6.2 Materials and Methods ...... 94 6.2.1 Cloning and production of Hbp truncation mutants ...... 94 6.2.2 Imaging and mechanical unfolding using AFM ...... 96 6.2.3 Image and force-distance curve analyses ...... 98 6.2.4 Computational estimate of unfolding parameters ...... 100 6.3 Results and Discussion ...... 102 6.3.1 Particle size distribution of the Hbp constructs ...... 102 6.3.2 Multi-step and multi-route unfolding pathways of Hbp . . . . . 104 6.3.3 Structure assignment of unfolding barriers ...... 108 6.3.4 Barriers coincide with stacked aromatic residues ...... 111 CONTENTS

6.3.5 Tracing the unfolding landscape of Hbp ...... 113 6.3.6 Hbp unfolds in domains ...... 114 6.4 Conclusion ...... 117

Bibliography 121

List of Publications 161

Summary 163

Nederlandse Samenvatting 165

Acknowledgments 167

Chapter 1

Introduction

The mechanics of biomolecules has become a new research frontier in the life sciences since the recognition that all aspects of living systems generate and respond to mechanical force. These responses are exhibited at different length scales from changes to the conformation of proteins, the shape of cells, the morphology of tissues, and the structure of organs. Other mechanical responses lead to chemical signal transduction in biophysical processes such as mechanosensation. Mechanical force generation in biological systems are also found in a wide range of length scales from molecular motors in cells, to extension and contraction in muscle tissues, where chemical energy in the form of ATP is converted to mechanical work. 1,2 Investigating how biological systems respond and generate mechanical forces is essential to understand the effect to its form and function. Studies have shown that mechanical force plays a vital role in cell differentiation, 3 tissue development and regeneration, 4 and tumorigenesis. 5 Understanding how biomolecules create, respond and transmit mechanical forces is indispensable to the development of synthetic materials for biological systems as well as the development of biomaterials for technological applications. Furthermore, mechanical experiments on biomolecules permit the application of physical principles and natural laws to biological processes and phenomena, thereby bridging the gaps between physics, chemistry and biology.

1.1 Single-molecule imaging

The study of biomolecules has advanced in the last three decades due to the development of single-molecule imaging and manipulation techniques. These techniques permit the imaging of molecules from cells, proteins and to DNA level. Single-molecule Single-molecule manipulation and force spectroscopy

Figure 1.1 Examples of super-resolution images. A) STED images of the mitochondria outer membrane, mitochondria matrix and superposition of the two 7 . B) Comparison between conventional ﬂuorescence (left panel) and STORM (right panel) images of microtubules (green) and clathrin- coated pits (red). 8 C) AFM images of DNA-RNAP (E. coli RNA Polymerase) complexes. The left images show stalled RNAPs in the presence of only three nucleotides: ATP, GTP and UTP. The right images show colliding RNAPs when CTP is added to the reaction mix. 9 (Copyrights © 2008, Nature Methods, Macmillan Publishers Ltd., © 2007, Science, AAAS)

manipulation has elucidated biological processes with unprecedented mechanical sensitivity in the application and detection of sub-nanonewton forces. In recent years, optic-based imaging techniques have reached sub-diffraction limits (50 to 200 nm) due to the development of super-resolution fluorescence microscopy such as STED (STimulated Emission Depletion), STORM (STochastic Optical Re- construction Microscopy), PALM (Photo-Activated Localization Microscopy), among others. 6 Sub-nanometer resolutions can be obtained using non-optical techniques such as electron microscopy, x-ray crystallography and Atomic Force Microscopy (AFM). However, the use of electron beams and x-rays demand harsh sample treatment such as freezing, and measurements are done in non-physiological conditions. On the other hand, AFM creates a topographical image of the surface and its surroundings (likened to the use of a blind man’s guide cane) in near-physiological conditions with sub-nanometer resolutions. Using these techniques, several biological systems have been imaged which were not thought accessible before as shown in figure 1.1.

1.2 Single-molecule manipulation and force spectroscopy

Single-molecule manipulation and force spectroscopy emerged as one of the powerful biophysical measurement techniques at the turn of the century. The use of these techniques led to fascinating insights into the physics (mechanics and dynamics) of biological systems and processes. The power of these techniques lies in the range of displacements and forces where they operate, i.e. biomolecules can be mechanically manipulated at nanometer-range displacements and applied with forces in the piconewton to nanonewton range. Thus, the force-distance relationship can be measured with very high precision and accuracy to determine the viscoelastic properties of DNA, proteins, viruses and cells; the strength of molecular bonds such as ligand

 1 Introduction

Figure 1.2 Single-molecule force spectroscopy techniques. 10 A) Dual-trap optical tweezers. The first trap tethers one end of the DNA while the second binds a polymerase. This experiment investigates the processivity and the forces involved during transcription. B) Magnetic tweezer. The permanent magnets produce a magnetic field gradient (dashed line) along the axial direction resulting to a force on the bead that is directed upwards towards the magnets. Rotation of the magnets (black circular arrow) results to a simultaneous rotation of the magnetic bead (red circular arrow), producing torque for unwinding the DNA. C) AFM. The ends of a polyprotein are attached to the tip and the surface. The protein is unfolded by vertically retracting the surface (mounted on a piezoelectric stage) and the force of unfolding is detected by the deflection of the cantilever. The measured force-distance curve displays the unfolding mechanics of the protein and its elasticity. (Copyright © 2008, Nature Methods, Macmillan Publishers Ltd).

and antibody binding and the unfolding pathways and energy landscapes of proteins. Figure 1.2 shows the schematic diagrams of the three most commonly used techniques: optical tweezers, magnetic tweezers and AFM. An optical tweezer or optical trap is created by a highly-focused laser beam and the use of dielectric particles. The interaction between the steep gradient of the optical field and the induced polarization of the dielectric particle creates a restoring force that is directed towards the focus. 11 For small displacements from the focus, the optical trap can be considered as a Hookean spring whose spring constant and stability is dependent on the steepness of the optical gradient, the laser power and po- larizability of the trapped particle. Biomolecules of interest can be manipulated and monitored by fixing them to a bead coated with a ligand or antibody, e.g. a biotiny- lated DNA molecule is tethered to streptavidin-coated beads as shown in figure 1.2A. The displacement and the force experienced by the trapped particle is simultaneously detected by imaging the bead using a position-sensitive detector or a high-resolution, high-sensitivity camera. 12,13 A magnetic tweezer uses magnets (permanent or electromagnet) to manipulate a magnetic particle. It exerts a force and torque to the magnetic particle, whose magnitudes depend on the magnetic field gradient and the particle’s magnetic sus- ceptibility. Three pairs of magnets allow full 3-dimensional position and rotation control of the particle through a feedback loop mechanism. 14 The relative simplicity of applying torque on a particle makes magnetic tweezers the ideal technique to study

 Physics of biomolecules

Table 1.1 Comparison of single-molecule force spectroscopy techniques Optical tweezer Magnetic tweezer AFM Stiﬀness (pN/nm) 10−3–1 10−6–10−3 10–105 Force range (pN) 10−1–102 10−3–102 10–104 Displacement range (nm) 10−1–105 10–104 1–104 Spatial resolution (nm) 0.1–2 5–10 0.5–1 Temporal resolution (ms) 0.1 10–100 1 Probe size (µm) 0.25–5 0.5–5 0.005–0.5 (tip) Special features multiple traps torque high-resolution imaging

the unwinding of naked DNA as shown in figure 1.2B and the motion of proteins and 15 molecular motors such as topoisomerases which unwind DNA and F0F1-ATPases which propel rotary motion such as that of bacterial flagella. 16 AFM (Atomic Force Microscopy) is composed of a cantilever with a very sharp tip, a laser to measure the cantilever’s deflection via the change in the laser’s reflec- tion onto a quadrant photodiode and a 3-dimensional piezoelectric stage to scan the sample surface and control the vertical motion of the cantilever. 17 The AFM is pri- marily an imaging technique that makes a topographical representation of the sample surface. As a contact microscope, it is exploited to manipulate and apply forces to biomolecules. Aside from high-resolution imaging of biomolecules, the AFM has a growing application in the viscoelastic measurements of nanostructures such as the indentation of viruses, vesicles, and cells 18 and unfolding of proteins 19 as shown in figure 1.2C. Table 1.1 shows a comparison of features of the different techniques. Optical and magnetic tweezers are most often used in elasticity measurements which require sub- piconewton to few piconewtons force range such as DNA and DNA-protein complexes, while the AFM is used when sub-nanonewtons to several nanonewtons force range is needed such as unfolding proteins and nanoindenting cells and viruses, as described in this dissertation.

1.3 Physics of biomolecules

The diﬀerence between single-molecule and ensemble or bulk experiments lies in the kind of averaging done when measuring the parameters of a system. Biomolecular properties and processes can be accessed in single-molecule experiments by following one molecule at a time, uncovering transient and intermediate phenomena, which are often obscured by population averaging in the case of ensemble experiments. Thus, single-molecule experiments can resolve individual state and individual species within

 1 Introduction the distributions of the thermodynamic and kinetic molecular properties measured in bulk experiments. In light of these experiments, physical principles can be directly applied to biology. The development of single-molecule experiments bridged the gap between physics and biology, paving the way to synergistic collaborations between physicists, biologists and mathematicians. Existing physical models were applied particularly from the ﬁeld of statistical physics and new theories were being developed to analyse these results, where Brownian motion and thermal ﬂuctuations dominate. 20

1.3.1 Elasticity of biopolymers, proteins and DNA

The isolation and identiﬁcation of the DNA (Deoxyribonucleic Acid) in 1869 by Friedrich Miescher 25 and the discovery of its double helical structure in 1953 by James Watson and Francis Crick 26 are heralded as one of the triumphs of science. The DNA molecule, aside from being the carrier of genetic information, plays a major role in the biological processes that constitute the central dogma of molecular biology, namely replication (producing two copies from one original DNA), transcription (producing mRNA from DNA) and translation (synthesis of proteins from mRNA). These processes involve bending, stretching, unwinding and unzipping the DNA and RNA; and subsequently the folding of proteins. Thus, investigating the response of these biomolecules to mechanical forces is important to understand the mechanics and dynamics of these processes. DNA and proteins can be regarded as polymers composed of monomeric subunits which are joined together. DNA is a polynucleotide made of 4 diﬀerent nucleotides: A, G, C, T which stand for Adenine, Guanine, Cytosine and Thymine, respectively;

Figure 1.3 A) FJC (freely-jointed chain) and WLC (worm-like chain) models of biopolymers under the influence of force. B) Force-extension curve when stretching double-stranded DNA. The eWLC model fits well the B and S forms of dsDNA while FJC fits well the ssDNA. C) Force-distance curve when unfolding a chain of titin immunoglobin domains, I27. The sawtooth pattern corresponds to unfolding each of the eight I27 proteins, which are fitted with the WLC model with Lp = 0.4 nm. Copyrights © 2012 Frontiers of Physics, 21 © 2009 Journal of Physics, 22 © 2011 Nature Physics, 23 © 2009 Journal of Physical Chemistry B. 24

 Physics of biomolecules while proteins are polypeptides composed of 20 different peptides or amino acids. In solution, DNA and proteins adopt configurations which minimizes the structural free energy. However, brownian motion and thermal fluctuations cause these biopolymers to form different conformations around a mean equilibrium conformation. Thus, the state of these biopolymers is akin to a problem related to random walk and diffusion whose solution can be derived from statistical mechanics. The influence of a force in extending or stretching these molecules constrain the number of accessible conformations, thereby reducing the configurational entropy of the system. For weak forces, the elastic response of these biopolymers is dominated by entropic as well as enthalpic effects, which correspond to the change in conformation and the rearranging of hydrogen bonds and π-stacking of nucleotide or peptide bases, respectively. The force-extension relationship has been successfully modeled by polymer physics using the freely-jointed chain (FJC) and the worm-like chain (WLC) models as shown in figure 1.3A. 21 The FJC or the ideal chain model is the simplest model to describe a polymer. It approximates the polymer as a chain of n segments with a characteristic length Lk (Kuhn length) and connected via freely-rotating joints. Thus it regards each segment as exhibiting a random walk with no interaction among the other segments. The derived force-distance relation for the FJC model is F = −kBT 2 , where kB, nLk T and nLk are the Boltzmann’s constant, temperature and stretched length of the polymer. On the other hand, the WLC model treats the polymer as a homogeneous elastic rod characterized by its contour length, Lc and persistence length, Lp which can be defined as the rod’s bending stiffness. The elastic response can be expressed as 27,28 k T 1 x −2 x 1 F = B 1 − + − . Lp 4 Lc Lc 4 The enthalpic contributions can be accounted for by assuming that each rod is extensible with a (normalized) spring constant K0 and substituting for xe = x − FLc/K0, yielding the extensible WLC (eWLC) model. 29 Figure 1.3B shows the force-distance curve in stretching double-stranded DNA (dsDNA) using optical tweezers. 22,23 Without tension, the DNA assumes the B-form conformation which is the double helix structure formed by Watson-Crick base pairing. Below 65 pN, the DNA’s response is dominated by entropic and enthalpic elasticity. The DNA then undergoes a conformation change to the S-form marked by a force plateau above 65 pN which causes unwinding of the helical structure and unstacking of the nucleotide bases. Both of these forms can be modeled by eWLC. Above 150 pN, base pairing breaks down and dsDNA becomes single-stranded DNA (ssDNA), which is best fitted by the FJC model.

 1 Introduction

Figure 1.3C shows the force-distance curve of a chain of eight I27 immunoglobulin domains of the human cardiac titin when stretched using an AFM. Titin is a protein found in striated cardiac and skeletal muscle tissues which acts as a molecular spring and is responsible for muscle elasticity. The unfolding of each protein can be modeled by WLC with Lp = 0.4 nm and shows an average unfolding force of 200 pN. These experiments provide quantitative information on the protein’s folding energy landscape which could allow predictions on the pathways, kinetics and dynamics of biological interactions and processes.

1.3.2 Elasticity of intra/extra-cellular ﬁlamentous networks

From a physics point of view, cells are similar to soap bubbles or other self-assembled amphiphilic assemblies such as micelles, which have membranes that isolate the inside contents from the outside environment. But in contrast to these structures, the cell membrane is connected to a cytoskeleton which is a dynamic and crowded network of cross-linked long-chain molecules. The cytoskeleton is the structural scaffold of the cell that provides its elasticity, shear rigidity and shape integrity. 32 In most cells, the cytoskeleton is composed of several filamentous macromolecules such as microtubules, actin and other intermediate/micro filaments. Figures 1.4A-C show that the cytoskeleton is responsible for the structure and form of the cell, and its dynamicity is due to the interplay of the different filaments. For example, cell migration is initiated by rapid actin polymerization at the cell’s front, which pushes and generates force for advancing the cell’s leading edge. The microtubules provide structural support and direct the cell movement by shrinking and growing towards the cell’s front edge. The microtubules may be viewed as tracks on which trains of molecular motors like kinesin and dynein walk through, as they supply and transport

Figure 1.4 A) Fluorescence microscopy of eukaryotic cells showing the nuclei (blue) and the cytoskeletal network such as actin ﬁlaments (red) and microtubules (green). B) AFM image of a cell. C) AFM image at high scanning force showing microtubules. D) AFM image of a microtubule with a kinesin motor walking on it. Sources and copyrights: http://rsb.info.nih.gov/ij/images/, © 2008 Cellular and Molecular Bioengineering, 30 © 2011 Biophysical Journal. 31

 Physics of biomolecules

Figure 1.5 Techniques in measuring the mechanics of the cytoskeletal matrix. A) Scanning electron microscopy image of a ﬁbroblast crawling on micropillars. Cell adhesion and traction forces are measured from the bending of the pillars. B) Schematic diagram of rheology experiments and model of the cytoskeletal matrix. C) Tracking the motion of microbeads anchored to the cytoskeleton. D) Optical tweezers or AFM experiments on cellular cytoskeleton. Copyrights © 2012 PNAS, 35 © 2008 SoftMatter, 42 © 2005 Nature Materials, 43 © 2012 PLOS One. 38

cargo such as vesicles and organelles throughout the cell. Figure 1.4D shows an AFM image of a kinesin motor walking along a microtubule. 31 The cell changes shape, generates force and regulates its elasticity by varying the self-assembly of these macromolecules. To understand how cells structurally behave and respond to its environment, several single-molecule studies were performed on the cytoskeletal matrix as shown in figure 1.5. In these experiments, the generation of force by the cytoskeletal network and their mechanical response to externally-applied force were studied in terms of its tensile, bending and shear elastic properties. Results from microrheology experiments have shown that the mechanics of cytoskeletal filaments are very similar and follows the same laws such as scale-free elastic properties (weakly-dependent on frequency) and the whole cytoskeleton is subject to constant prestress. 33 Techniques such as micropillars have measured the tension and force gra- dients from an actin network 34 and found that cell migration and mechanosensing are also dependent on the substrate stiffness. 35 Using optical tweezers and AFM, the elastic moduli of individual filaments were measured such as that of spectrin, microtubules and actin. 36–38 Aside from intracellular filamentous networks, elasticity measurements were also performed on extracellular matrix especially those involved in connective tissues such as collagen and elastin as well as amyloid fibrils which are insoluble protein aggregates found in Alzheimer’s and Parkinson’s diseases. 39 Com- plementary to these experiments, theoretical models have been developed such as the extension of the classical WLC model called glassy WLC (gWLC), tensegrity and network models (Hookean spring, passive/active cable and semi-flexible polymer) 40,41 of the cytoskeletal as well as the extracellular matrix.

 1 Introduction

1.3.3 Elasticity of viruses, vesicles and cells

Intact structural integrity and stable mechanical property are one of the essential criteria for biomolecules to carry out their physiological functions accurately. Factors that otherwise disrupt these have led to diseases or apoptosis for cells and reduced infectivity for viruses. For example, sickle cell disease is caused by abnormality in the shape and elastic properties of the red blood cells which causes jamming and aggregation in veins and arteries, 45 breast cancer cells can be distinguished from normal ones by their reduced stiffness and high-deformability , 46 and matured fully- infective human immunodeficiency virus (HIV) are softer than immature less-infective ones. 47 To measure the mechanical properties of different biomolecular structures, different methods are employed. Nanoindentation using the AFM with sharp or blunt tips (by using tipless cantilevers or attaching a bead on the tip) is the most common method used. 44 In these experiments, the forces applied by the tip and the amount of indentation experienced by the biological structures are recorded with piconewton and nanometer resolutions. The resulting force-distance curve provides a measure of the structure’s elasticity and mechanical strength as shown in figure 1.6. In this example, the nanoindentation on Herpes simplex virus capsids reveals its elasticity and mechanical strength from the analysis of the force-distance curves. For instance, the slope of the linear part in figure figure 1.6B determines the spring constant of the virus, (k) and the point where a sharp drop in force occurs, (Fc) provides a measure of its mechanical strength. Using this kind of experiment, the mechanical properties of different virus species have been measured. 18 The same method has also been applied to vesicles and cells. 48

Figure 1.6 Schematic diagram of AFM nanoindentation experiment and the recorded force-distance curve. A) The piezo stage approaches the tip, but has not yet touched the sample thus the recorded force is still zero. B) The tip starts indenting the sample. The force exerted by the tip is proportional to the bending of the cantilever, which is recorded by the change in the direction of the laser’s reﬂection onto the photodiode. The resulting force-distance curve describes the mechanical properties of the sample, such as the spring constant, k and critical force, Fc. The AFM images of the virus capsid (Herpes simplex) show an intact capsid with visible pentons and hexons before indentation and a broken capsid after indentation. (Copyright © 2010, Nature Physics, Macmillan Publishers Ltd 44)

 Physics of biomolecules

Along with the development of these experiments, theoretical models and simulations have emerged, including the application of the principles from continuum elasticity theory. 49 For instance, it has been found from nanoindentation experiments on nanometer-sized bacteriophages and viruses with icosahedral symmetry that their elastic properties follow the thin-shell elasticity theory. 44 This is the same principle applied by engineers to thin-walled macrostructures used in airplanes or cars. In this model, the force applied is proportional to the indentation δ of the particles. However, when the inside contents are highly-viscous such as in vesicles and cells, or for small viruses where the shell thickness is comparable to the radius, the thin-shell approximation breaks down. In these cases, the Hertzian model is applied where the force scales with δ3/2. Finite element models are also used in simulations on indentation of viruses and cells, which takes into account the actual sizes of the particles and the indenter (e.g. AFM tip). 50,51 These simulations can capture the biophysical processes that occur during indentation which are currently inaccessible to experiments. The advances made in the study of biomolecules - their properties and the processes they are involved in - have been pushed forward in the last decade through the use of novel single-molecule biophysical techniques. These techniques enable us to see deeper into the basic molecules that make up most of living systems such as DNA and proteins, and allow us to observe the primary processes occurring in the simplest unit of life such as cells and viruses. It is fair to say that these techniques have revolutionized the way we study biological structures and processes - by quantifying biophysical properties and interactions and applying physical principles to biological processes. These experimental and theoretical investigations enable the direct application of physics in biology. This provides a striking demonstration that the laws of nature hold reign over all forms of matter from non-living to living systems - the relatively recent realization that the same laws can be invoked to seamlessly explain structures from the astronomical to the microbial level.

 1 Introduction

1.4 Outline of this thesis

Atomic force microscopy (AFM) is the main technique used to perform imaging and force spectroscopy on the biomolecules discussed in this dissertation. DNA, proteins, nanocolloids and viruses were probed and imaged at high spatial resolution revealing their size, conformation and structure; which are otherwise not possible with optic- based microscopy. Forces at the nanonewton range are applied on viruses and proteins by pushing and pulling with the AFM tip. Viral capsids were pushed with the AFM tip to reveal their mechanical properties and structural strength. Proteins were fixed on the surface and their other ends were fished out by the AFM tip, and pulled to unravel their elastic properties and unfolding pathways. In chapter 2, the application of AFM is reviewed in probing the form and function of biomolecules, proteins and supramolecular assemblies by combined microscopy and force spectroscopy measurements in air as well as in liquid conditions. The capabilities of AFM are categorized into three method areas: high-resolution imaging, nanoindentation experiments and unfolding experiments. As a microscope, the AFM is used to study the protein’s form; and as a force spectroscope, it is used to study the protein’s function by measuring its elasticity and mechanics. Thus, with the AFM results, a protein’s particular form can be directly associated to a specific function. In chapter 3, a collection of experiments using high-resolution AFM imaging are discussed. Mitochondrial DNA molecules were imaged with sub-nanometer resolution and the effect of DNA-binding protein TFAM (Transcription Factor A in Mitochon- dria) on the compaction of DNA was visualized. This was further quantified by measuring the traceable contour length and end-to-end distance of the DNA-TFAM complexes. The results show that at physiological concentrations, fully-compacted and nearly-naked DNA coexist which suggest that TFAM has a regulatory role in the number of DNA molecules available for replication and transcription. The second experiment is the imaging of nanocolloidal albumin and investigation of the effect of radioisotope labeling with 99mTc and 89Zr, and fluorescent dye IRDye 800CW. The results show that the size of the 20-nm nanocolloids remains the same and no aggregation occurs. These labeling techniques open up additional techniques such as PET scan and fluorescence microscopy for imaging tumors and cancer cells. The third experiment is the imaging of Norwalk virus-like particles (NVLP) and the investigation of the reversible assembly and disassembly of NVLPs at different pH conditions. The results show that NVLPs form 38-nm particles with T=3 icosahedral symmetry at acidic conditions (2

 Outline of this thesis enabled the virus to withstand the fluctuating pH conditions during its oral-to-fecal route of infection. In chapter 4, the advances in the study of virus mechanics using AFM are reviewed. The strength of AFM as an imaging tool is manifested in the application to single viral capsids that were imaged at unprecedented sub-nanometer resolutions under physiological conditions. Individual proteins that make up the viral capsids (pentons and hexons), genome release and viral budding from cells were clearly visualized. Moreover, using AFM as a force tool the mechanics and structural strength of the capsids were measured. For instance, the effect of genome packaging, maturation and mutations on the elastic properties were investigated and how they affect the infection cycle of the virus. This review highlights the emerging role of the AFM as a single-molecule biophysical technique aimed at studying the elasticity of biomolecular structures from viruses, vesicles to cells. In chapter 5, nanoindentation experiments on Norwalk virus-like particles are discussed. Norwalk virus is a human pathogen that causes gastroenteritis or stomach flu. It is a non-enveloped virus which means that its genome is encapsulated and protected alone by its 38-nm capsid. The capsid is composed of a shell domain and a protruding domain with radially-extending protrusions containing the virus’s receptors to the host cell. Using a combined AFM imaging and force spectroscopy approach, the influence of the protruding domain on the overall structural stability of the virus was investigated by comparing the mechanical properties of the particles with and without protrusions (wild type and mutant, respectively). The results show that the wild type particles exhibit stiffer yet brittle behavior while the mutant particles exhibit plastic deformation. The size and elasticity measurements indicate that the presence of the protruding domain introduces a prestress in the viral capsid which is responsible for the increase in the stiffness and brittleness of the wild type particle. The results suggest that the virus is in continuous prestress condition which enhances its structural stability during the infection cycle and that the prestress might have facilitated the genome release during the virus’s docking at the host cell’s surface. In the last chapter, the unfolding experiments on Haemoglobin protease (Hbp) are discussed. Hbp is an autotransporter protein secreted by Escherichia coli. It is involved in the symbiotic relationship between E. coli and B. fragilis, and is used to scavenge heme for both species by degrading haemoglobin from the host, thereby causing peritonitis or stomach ulcer. The secretion of autotransporters, including Hbp from the bacterium’s outer membrane (OM) is an ATP- and proton gradient- independent process. It is hypothesized that their translocation is facilitated by sequential folding of the protein at the surface of the OM and in the process pulls itself out of the membrane. This hypothesis was investigated, albeit in reverse, by

 1 Introduction unfolding Hbp and exploring its unfolding pathways. This was realized by ﬁxing one end of the protein (functionalized with cysteines on both ends) on a gold-coated surface and ﬁshing out the other end with a gold-coated AFM tip. The results show that Hbp unfolds in a sequential multi-step and multi-route fashion along stable unfolding barriers. These barriers are found to correspond to the unfolding of each of the helices that make up the β-helical structure of the protein. We therefore deduce that the folding of these β-helices creates a stable conformation in the protein that may provide the energy to pull and translocate the protein. These results support the folding-driven transport model of Hbp, where the attractive interaction between the stacked aromatic residues drives the processive folding of the β-helical backbone and initiates the translocation process that eventually pulls out the rest of the protein across the outer membrane. The content of this dissertation describes several experiments that elucidate the elastic properties of a variety of biomolecules. The mechanical experiments on these molecules provide access to processes that enables quantitative measurement of their properties and interactions. The application of physical principles to these biological processes furthers our understanding of what living systems are made of, how they work, and why they behave like they do.



Chapter 2

Sampling protein form and function with the atomic force microscope

Published as Molecular & Cellular Proteomics 9 1678-1688, (2010). 52

Abstract

To study the structure, function and interactions of proteins a plethora of techniques is available. Many techniques sample such parameters in non-physiological environments (e.g. in air, ice or vacuum). Atomic Force Microscopy (AFM), however, is a powerful biophysical technique that can probe these parameters under physiological buﬀer conditions. With the AFM in these conditions it is possible to obtain images of biological structures without requiring labeling as well as following dynamic processes in real-time. Furthermore, by operating in force spectroscopy mode, it can probe intra-molecular interactions and binding strengths. In structural biology it has proven its ability to image proteins and protein conformational changes at sub-molecular resolution and in proteomics it is developing as a tool to map surface proteomes and to study protein function by force spectroscopy methods. The power of AFM to combine studies of protein form and of protein function enables bridging various research ﬁelds to come to a comprehensive, molecular level picture of biological processes. We review the use of AFM imaging and force spectroscopy techniques and discuss the major advances of these experiments in further understanding form and function of proteins at the nano-scale in physiologically relevant environments. Introduction

2.1 Introduction

In order to understand biological processes at the molecular level it is essential to identify the involved proteins and proteinaceous assemblies, to characterize their structure and function and to unravel their interplay with other proteins and molecules. 53 Techniques like X-ray crystallography, electron microscopy, nuclear magnetic resonance spectroscopy and mass spectrometry have contributed massively to elucidate such protein properties. These techniques can easily sample the properties of a large ensemble of proteins; however they require subjecting the sample to harsh treatments such as drying, crystallising or vaporising in vacuum, thereby limiting the range of measurable dynamical properties of the sample. One powerful method that permits the investigation of molecules in their native physiological buffer condition is Atomic Force Microscopy (AFM). 17 An AFM is a microscope and force spectrome- ter at the same time. The imaging resolution of the AFM is comparable to that of electron microscopes and it has the special capability to image samples in a variety of environments such as in vacuum, air or liquid, which therefore enables studying biological specimens in their native environments (i.e. in buffer solutions). 54,55 In addition, its ability to “touch” the sample gives it the advantage to manipulate single particles/molecules and probe their mechanical properties. 18,56–58 However, AFM force spectroscopy is currently a technique with rather fast pulling and pushing speeds, thereby often operating out of equilibrium conditions. Improvements with ultrastable AFM are underway to tackle this problem with promising results. 59,60 Furthermore, the AFM is not well suited to apply and resolve forces at the single piconewton range due to large-size tips and relatively stiff cantilevers. The issue of non-specificity of the tip interaction with the sample is also of concern, especially in pulling experiments which require the capability to accurately recognise and select the appropriate molecule or point of interest. The current introduction of carbon nanotube tips can address the former issue 61,62 while techniques in chemical function- alisation can provide directed tip specificity and recognition capability, 63–68 thereby further improving and widening the applicability of AFM in the future. In addition, the coupling of AFM to fluorescence microscopes further enhances its versatility by adding (single-molecule) fluorescent imaging to the AFM imaging capability 69–71 and the development of high-speed systems makes it possible for AFM to probe fast dynamics of various biological processes. 72–76 The applicability of AFM in proteomics is diverse and includes the characterisation of the cell surface proteome (see ref. 77 for a recent review), label-free detection and counting of single proteins, 78,79 and force spectroscopy measurements of binding and unbinding events. 80,81 In structural biology, AFM has shown to be a powerful tool for high-resolution imaging of proteins in near-native conditions, 55,57 and structural

 2 Sampling protein form and function study of supramolecular assemblies like protein filaments and viruses by nanoindentation methods. 82,83 These experiments show the potential of AFM to study both “form” and “function” of proteins thereby resolving questions in proteomics and structural biology quasi-simultaneously. In the following we will explain the principles of atomic force microscopy and its different operation modes and finally discuss examples of imaging, nanoindentation and protein (un)binding and unfolding studies using AFM.

2.2 Principles of atomic force microscopy

The AFM is a member of the Scanning Probe Microscopy (SPM) techniques which utilises a probing tip that scans the surface of a sample. The measured interaction between the sample and the probe (e.g. current in Scanning Tunneling Microscope (STM) or force in AFM) renders a three-dimensional image of the sample surface. In AFM, the probe is a sharp tip with a typical radius of about 1 to 20 nm and mounted on a cantilever. The interaction between the sample and the tip is recorded by the bending of the cantilever as shown in figure 2.1. The deflection of the cantilever is monitored by the change in direction of the reflected light (laser) which is recorded by a position-sensitive detector (quadrant photodiode). Soft cantilevers with spring constants of about 0.01 to 0.1 N/m can sense forces as low as a few piconewtons; while modern piezoelectric scanners can translate the sample or tip in x-,y- and z-

Figure 2.1 Schematic diagram of an atomic force microscope in liquid. The cantilever is mounted on glass to facilitate access of the laser while probing a sample in a liquid cell. Light is reﬂected from the cantilever and detected by the quadrant photodiode. The photodiode signal is used to instruct the piezoelectric scanner to translate the x-,y- and z-positions of the sample through a feedback loop mechanism.

 Principles of atomic force microscopy directions with sub-nanometer resolutions. This combined functionality provides the AFM with the capability of rendering 3D images of the sample with atomic resolution and manipulating them at single molecular level with nano-scale forces. 84–87 As a microscope, there are various modes available to construct an image of the sample. Choosing the appropriate imaging mode and parameters influence the amount of contact force between the sample and the probe tip. 88–90 The different operation modes can be divided into static and dynamic categories. 91–93 In static imaging mode, the sample is brought into physical contact with the tip until their interaction reaches a condition where it satisfies a pre-defined parameter setting, such as the specified z-position of the sample (constant height mode) or the amount of interaction force between the tip and sample (constant force mode). In constant height mode, the sample is moved in x-y direction while maintaining its z-position. The surface or height information of the sample is reconstructed from the deflection of the cantilever, thus the measured contact force. However, since the distance between tip and sample is fixed, lateral dragging of the sample during imaging is quite common. In constant force mode, it is the imaging force which is set on a fixed value. The sample is brought in contact with the tip by moving the piezoelectric scanner in the z-direction until the contact force reaches the set value. The height of the sample at that position is then determined by how much the sample was moved in z. The tip is then moved to the next x-y position and using a feedback mechanism, will retract or approach the sample until the set contact force is obtained. The constant force mode provides a careful and cautious way of handling the sample by setting the contact force to a minimum. However, the cantilever normally moves laterally in a continuous manner and therefore the retraction or approach during feedback can still impart considerable dragging or damage during these lateral movements, especially in imaging soft biological samples. An alternative way of performing the constant force imaging mode is called jumping mode. This mode provides a safer scheme by retracting the tip to a defined height away from the sample before moving on to the next lateral x-y position. 94 With this mode, there is less probability of dragging the sample laterally because each time it moves to another position it is retracted first away from the sample surface which could be set such that it is above the highest bump or corrugation. In addition, because it also operates in contact force mode, the tip-sample interaction force can be pre-set to the lowest possible value thereby preventing damage to the sample. In dynamic imaging mode, the tip is made to oscillate while it approaches the sample. The change in the oscillation amplitude or frequency during approach is used to construct its surface information. In this mode, the tip can be touching intermittently with the sample (tapping mode) or not touching the sample at all

 2 Sampling protein form and function

(non-contact dynamic mode). In tapping mode, the contact between the tip and sample is minimised by oscillating the cantilever, so that the tip bounces up and down and therefore only ‘taps’ the surface as it moves around the sample. 95 The cantilever is made to oscillate lower than the natural resonant frequency; but when it approaches the sample, the resonant frequency decreases due to damping effects. The result is an amplitude increase because the driving frequency is now closer to the new resonance. Finally, the oscillation amplitude reduces again when the tip hits the sample. The monitored change in the oscillation amplitude is used as the feedback signal for constructing the image of the sample surface. In non-contact dynamic mode, the tip is never in contact with the sample. To implement this, the cantilever is made to oscillate at a frequency higher than resonance. As it approaches the sample, it’s amplitude of oscillation decreases because of damping. The closer it is to the sample, the lower is its amplitude of oscillation. The tip is prevented from coming too close to the sample by setting a limit on the lowest possible oscillation amplitude. As with tapping mode, this parameter is used as the feedback signal for image construction. The dynamic modes provide the option of little physical contact with the sample during imaging. In force spectroscopy, the AFM is operating in the force-distance measurement mode. 96,97 In this mode, the force is now specifically monitored and recorded as the tip is brought into contact (e.g. pushing/indentation experiments) or out of contact from the sample (e.g. unbinding and stretching experiments) while simultaneously recording the amount of distance of approach or retraction of the tip. In this mode, no lateral movement is performed during the approach and retraction cycles. Typical applications of spectroscopy mode in protein studies are in pushing or nanoindentation of viruses and microtubules to probe their mechanical properties 37,82,83 and in pulling or stretching experiments to measure the (un)binding strength of two protein molecules 98–101 or measuring the molecular bonds of the different structural elements comprising large molecules such as muscle protein titin, 102,103 ubiquitin 104,105 and membrane protein bacteriorhodopsins. 106–108 In these experiments, the plot of the measured force against the deformation of the sample (force-distance curve) provides information on the material properties such as the spring constants, Young’s moduli, and the binding and unbinding, stretching and breaking forces of the probed structures. A typical force-distance curve generated by nanoindenting a virus particle is depicted in figure 2.2. In the beginning, the tip is far from the sample thus the measured interaction force between them is zero. At the instance the tip touches the sample surface, the cantilever starts to bend and the force-distance curve starts to rise as the tip pushes further on the sample. When the tip is retracted, the retraction curve may

 Principles of atomic force microscopy simply trace back the curve during approach (in the case of a hard surface) or retraces a completely different curve depending on the properties of the sample. 86,97,109,110 As illustrated in figure 2.2, the virus particle has exceeded its elastic limits during approach at about 0.8 nN and thereafter suffered irreversible deformation, maybe breakage. During retraction, the tip did not feel an intact virus particle anymore and the retraction curve resembled that of the glass curve. To ensure that the measured parameters were accurate and precise, the piezoelectric scanner, spring constant of the cantilever and photodiode signal must be calibrated. The scanner can be calibrated by imaging surfaces or microgrids with known dimensions. For the cantilever calibration, there are three different methods in measuring its spring constant: dimensional, static and dynamic methods. Di- mensional methods calculate the spring constant based on the cantilever’s physical dimensions and material properties, 111,112 the static method is based on the cantilever’s deflection in response to a known force, 113,114 while the dynamic method measures the resonant frequency and Q-factor by oscillating the cantilever. 115–117 Most AFM employ the beam bounce method to measure the cantilever deflection using a photodiode. Thus, an a priori calibration procedure is important to convert the photodiode signal into the actual force applied by the tip. Force calibration is

Figure 2.2 Example of a force-distance curve. The plot shows a force-distance curve when the AFM tip is pushed onto a hard surface (e.g. the glass substrate) and pushing on a supramolecular proteinaceous structure. During approach, as the tip is still far from the sample, the measured force from the cantilever is zero. As the tip pushes on the sample, the force increases with the amount of distance pushed. In pushing hard samples like glass, this relationship is linear, showing the elastic property of the cantilever (glass curve). In this case, the retraction curve retraces the approach curve. For deformable protein samples, in this example a viral capsid in buﬀer solution, the curve may not be completely linear and sudden drops of the force can occur. This can be interpreted as buckling or breakage in the sample. In this example the retraction curve is very diﬀerent and implies that the indented particle was irreversibly deformed during approach (for at least the time scale of the experiment and probably permanently).

 2 Sampling protein form and function usually implemented by pushing the tip on a hard surface such as a mica or glass substrate. This curve measures the deflection due to the cantilever alone, referred to as the glass curve in figure 2.2. When pushing or pulling a sample, its actual indentation or stretching can be measured by subtracting the cantilever’s deflection from the recorded force-distance curve, as demonstrated by the line with double arrows in figure 2.2.

2.3 High resolution imaging

Since its development as an imaging technique, AFM has become a wide-spread tool in academic and industry settings for solid state and materials research. 17,89 Being a contact microscope, it does not suffer from diffraction limits as optical microscopes do, thereby providing it with the capability to reach atomic resolutions. Furthermore, specific tip-sample interactions such as electrostatic and magnetic forces were further exploited to characterise material properties of solid state and polymer samples. The application of AFM has since then encompassed other areas of research. In this section, we are going to discuss the advances in imaging and force spectroscopy in the study of proteins and other biological samples.

100 nm

Figure 2.3 Sample AFM image of Noroviruses in liquid. On the left is a top view image showing several viral particles. In this view, structural details of the particles (cup-like depressions) can already be recognized. On the right is a zoomed-in side view image on one of the particles shown in a 3D representation. The cross sectional profile shows the apparent overestimation of the lateral dimensions of the sphere-like particle, which results from the tip-convolution effect, inherent to AFM imaging. 118,119 However, the diameter can be accurately determined by measuring the height. The z-profile indicates a maximum height of about 38 nm, which is in close agreement to the diameter determined by electron microscopy and x-ray crystallography. 120,121 The images were recorded using jumping mode AFM (see ref. 122 for full experimental details, with the sole difference that these images were taken in sodium phosphate buffer).

 High resolution imaging

AFM can render images of samples in different ways based on the used measurement parameters. The most common imaging is topographic, in this case the image of the sample surface is reconstructed from the z-movement of the piezoelectric scanner. An intuitive image can be also be derived from the error signal, i.e. the input to the feedback loop system, which can be interpreted as the first-order derivative of the topography image. This image enhances the fine details of the sample at the expense of the coarse features. The side-to-side difference signal of the photodiode which measures the frictional force experienced by the tip as it scans the sample 123–125 and the phase information in dynamic mode AFM 91,126,127 render again different aspects of the sample. These images reveal material characteristics of the sample such as viscoelastic and electrostatic properties, however this information is quite complex and hard to quantify, thus these images should be carefully analysed and in most cases only serve as intuitive supplements. The capability of AFM to image in liquids caught the attention of the biology community and has since then been added to the palette of available techniques to image biological samples. 128–132 A myriad of examples ranging from imaging cells, bacteria and viruses to isolated molecules such as nucleic acids (DNA, RNA) and individual proteins, show the AFM’s flexibility to image and manipulate (fragile) samples in their native state without destroying them. One of the first biological specimens that were imaged in aqueous solutions is the human blood protein fibrinogen which was followed in real-time during its clotting process. 133 AFM images of the fibrinogen before and after the addition of thrombin clearly show how it transforms the protein to fibrin monomers and catal- yses the spontaneous polymerisation of the monomers to chains and the subsequent cross-linking of these chains to form intricate fibrin nets. These fibrin nets play vital roles in biological processes such as healing wounds and producing blood clots that cause heart attacks and strokes. The first AFM images of isolated viruses and phages were in air which were initially used for tip calibration and reference purposes. 18,119 Around the same time cell-associated viruses in liquid were also imaged. 134 Since then a series of experiments have been performed to improve the quality of AFM images of viruses and viral subunits 18,135 (example shown in Fig. 2.3). Nowadays it is possible to image specific substructures like bacteriophage connectors and tails, 127,136,137 individual capsomeres of plant and human viruses, 135,138 new viruses budding from cells 18,125,134,139,140 and expelled viral genomes 141–143 under liquid conditions. Visualisation of fragile proteins in their native state such as mitochondrial as well as photosynthetic membrane proteins (e.g. bacteriorhodopsin) 144,145 were also done during the early developments of AFM. The surface topography of bacteriorhodopsin was resolved down to 1 nm 146 and since then vertical resolution of up to ∼1 Åwas attained as demonstrated by the imaging of the photosynthetic core complex in na-

 2 Sampling protein form and function tive Rhodopseudomonas viridis membranes. 147 The images of the latter experiment show a single reaction centre encircled by an ellipsoid of 16 light-harvesting subunits. Furthermore, by applying loading forces to the tip, different protein complexes were nanodissected out of the photosynthetic core and by imaging afterwards, the structural changes and rearrangement of the protein assembly were observed. 147 AFM images were also used to measure and probe material properties such as persistence lengths of DNA 148–151 and structure and defects of lipid films. 152 Aside from visualisation of biological objects, topographic AFM images were also used for surface roughness analyses to study the biocompatibility of scaffolds by correlating surface characteristics with (stem) cell proliferation 153,154 or the adsorbed proteome repertoire. 155 Furthermore, time-lapsed AFM imaging enabled tracking the motion of proteins at sub-nanometer resolutions, 156,157 real-time probing of protein-protein interactions such as the binding and dissociation of chaperonin protein GroES to individual GroEL proteins, 132,158,159 observing the disassembly dynamics of microtubules 160 and diffusion and transcription of E. coli RNA polymerase. 161,162

2.4 Indentation experiments

Aside from topographic imaging, AFM is also widely used for single molecule manipulation and for the application of forces in force-distance measurement mode. In these measurements, which can be categorised into pushing or pulling experiments, the application of force by the tip is well-controlled while the indentation or stretching of the sample is precisely monitored. The capability of AFM to physically sense interactions allows the study of inter- and intramolecular forces within the sample. 163 In biological applications, experiments range from determining the mechanics and structure of cells, supramolecular assemblies and single molecules as exempliﬁed by: (i) nanoindentation of cells, 46,164–166 microtubules, 37,167,168 viruses 83,169 and globular proteins, 170,171 as well as probing molecular bond strengths as demonstrated by (ii) unbinding, stretching and unfolding of receptor/ligands, 172–174 proteins 100,101,175 and nucleic acids. 86,149,176–178 In addition, inherent tip-sample interaction forces which include adhesion, capillary, Van der Waals and electrostatic forces can also be used to reveal material and chemical properties of the sample. Nanoindentation experiments are performed by employing the tip to push on the sample and recording the force-distance curve. In this case, the recorded deﬂec- tion of the cantilever depends on the viscoelastic property of the sample. 82,109,179 When the measured stress (force) and strain (deformation) are linearly-related, the material is said to be linearly elastic and it will normally regain its original form after relaxation. Nanoindentation experiments can measure local or global elasticity

 Indentation experiments and can be expressed in terms of the Young’s (or elastic) modulus. Furthermore, by applying higher forces, the AFM can also probe the structural strength of the sample from the force where it starts to irreversibly deform or break. For instance, in a subfield of the emerging research domain of Physical Virology (i.e studies of material properties of viruses, forces involved in viral DNA packaging, genome ejection mechanisms etc. 180,181), the AFM is used to analyse the mechanical structure of viral particles. These nanoindentation experiments have shown that the Young’s moduli between viruses can differ significantly, 83 but that one can correlate those differences with the density of the protein packing in the capsid. 182 Viruses with a loose protein packing, like Cowpea Chlorotic Mosaic Virus (CCMV) and Hepatitis B Virus (HBV) have a Young’s modulus below 0.5 GPa, 183,184 whereas viruses with a capsid structure that exhibit more extensive protein-protein contacts, like the bacteriophages Φ29 and λ and the Minute Virus of Mice (MVM) possess Young’s moduli of at least 1.0 GPa. 169,185,186 Remarkably the differences in Young’s moduli also seem to reflect the packaging mode for genome encapsidation. Viruses that self-assemble around their genome (CCMV and HBV) do not need a rigid shell as it is unlikely that high pressures will be exerted onto their inner walls by the genome. However, when a packaging motor is used to encapsidate the genome to pre-formed capsids, such as phage Φ29, phage λ, MVM (see also 83) and Herpes Simplex Virus type 1 (HSV1), 138,187 higher internal forces will generally be exerted onto the capsid walls. The combination of AFM nanoindentation measurements with theoretical and modelling approaches has provided additional information on the studied viral capsids like the particle’s Young’s modulus and the deformation behaviour under high forces (for a review on modelling approaches of viral capsids see ref. 188). Finite element simulations and continuum elastic theory have been applied to understand the experimental results of for instance phage Φ29, 169,188,189 CCMV, 50,183,188–190 MVM, 185 HBV, 51 Murine Leukemia Virus, 191 Human Immunodeficiency Virus 47 and Tobacco Mosaic Virus. 192 As these modelling approaches are generally completely reversible, they cannot explain in detail irreversible processes like capsid breakage at large de- formations. To do this, one needs to resort to other modelling approaches like for instance molecular dynamics (MD) simulations. 182 Using such models permit interpretation of the experimental results at the protein level, a major step forward in our understanding of the mechanics of protein shells which is not possible with experimental techniques alone. AFM nanoindentation experiments on other regular supramolecular assemblies, like microtubules, have been performed to study the material properties of these self-assembling protein tubes. 37,167,168,193,194 Noteworthy is a nanoindentation study of the decoration of microtubules with the microtubule-associated protein tau. 194

 2 Sampling protein form and function

In this study it was shown how tau binds in a 1 nm thick layer around the microtubules, presumably along the protofilament tops. The radial spring constant of the microtubules was unchanged upon tau binding, however, the resistance against failure increased slightly. This experiment shows how form and function of protein assemblies can be studied using AFM. In another example of AFM studies of microtubules, not only the Young’s modulus but also the shear modulus of microtubuli was determined. These measurements were performed by indenting microtubules that were lying over holes in the supporting substrate. 168,195 This approach was further developed for mechanical studies of another cytoskeletal protein polymer, vimentin intermediate filaments. 196 A length dependent bending modulus was inferred from the data, suggesting axial sliding between the subunits of intermediate filaments. In a similar approach the mechanical properties of elastic fibers and collagen fibrils were probed by freely suspending them over microchannels in a solid support. 197–199 The two order of magnitude difference between the Young’s and shear modulus of type I collagen fibrils, which was determined, confirmed their mechanical anisotropy. As with viral capsids, the studies of protein filaments have also benefited from modelling approaches like finite element and MD simulations to determine for instance accurate values of the Young’s modulus. 195,200,201 The mechanical properties of selected globular proteins were also studied by AFM nanoindentation as discussed in. 82,171 Examples include force curves on lysozyme molecules 170 and on bovine carbonic anhydrase II, a zinc metalloprotein. 202 The latter experiments describe the change in mechanics of this zinc metalloprotein under increasing denaturing conditions, revealing a significant softening of the molecules at the onset of denaturation. In another experiment by the same group the compression of surface adsorbed green fluorescent proteins was studied. They showed a reversible quenching of the fluorescence around a force of 5 nN. 203,204 Whereas AFM nanoindentation experiments on supramolecular assemblies have provided insight into the mechanics and structure of a variety of biological complexes, such experiments on single proteins are still very scarce. This is probably due to the difficulty to interpret single-molecule compression results, but this situation might change with further developments in modelling approaches. 82 Contrasting with the small amount of “pushing” experiments on single molecules, there has been a wealth of “pulling” experiments as we will discuss now.

 Unfolding experiments

2.5 Unfolding experiments

The interactions between isolated pairs of molecules, such as that of protein-protein or receptor-ligand bonding, 65,80,101,110,205,206 have been probed by AFM force spectroscopy as well as the distribution and interaction forces of proteins on cell surfaces. 65,77 These experiments generally require functionalised AFM tips and substrates, and immobilising either of the partners on the tip or on the surface while taking force- distance curves. From these experiments, binding strengths of structural and chemical bonds and their breaking forces were measured. The binding strength between biotin and (strept)avidin for fixed loading rates (the ratio of pulling force or piezo- movement over time) was probed by various groups and averaged interaction forces of tens to hundreds of piconewtons were found. 101,172,173,207 To study the energy landscape of bond strengths however, one should probe the interaction force for different loading rates. 208–210 Doing so for the biotin-streptavidin bond shows that bond strengths increase with increasing loading rates. 174 Next to receptor-ligand interactions, antibody-antigen interactions were also probed by AFM force spectroscopy. The unbinding force between human serum albumin and a polyclonal antibody to this molecule was found to be ∼250 pN. 63 The interaction of intercellular adhesion molecule-1 with a specific antibody was probed in a combined force spectroscopy and adhesion imaging approach. 211 Here it was shown that topography information was correlated with specific molecular recognition. This is nice example of how form and function of proteins can be simultaneously studied with one and the same technique. Interaction between other proteins have also been probed, for instance, the binding strength between the motor protein myosin and its molecular track actin 212 and the interaction force between Escherichia Coli chaperonin GroEL and two different substrates. 109 In the latter study it was found that the binding strengths are lower in the presence of ATP and that denatured proteins bind stronger to GroEL than proteins in a native-like state. Finally, intermolecular forces in nucleic acids were probed by AFM force spectroscopy in an approach to pull apart two single strands of DNA, using complementary oligonucleotides consisting of 20 bases. 176 In AFM stretching and unfolding experiments of proteins, single- or multi-domain proteins are functionally attached from one end to an AFM tip and the other end to the surface. Then the tip is pulled away from the surface, stretching and elongating each domain of the protein. In this scheme, at certain force ranges, a molecular bond or a structural element unfolds resulting in a sudden increase of the contour length and drop of the force. Depending on the protein, this could be considered an elastic extension or an irreversible rupture which depends on the pulling speed, temperature and solvent conditions, among other factors. 80,163 The first publication among these experiments was the repeated stretching of the giant muscle protein

 2 Sampling protein form and function titin, exhibiting a reversible unfolding of the individual immunoglobulin domains. 175 Furthermore, it was shown that one can mechanically induce the binding of ATP to titin kinase which means that titin kinase acts as a biological force sensor. 213 Stretching experiments on the fluorescent proteins EYFP 214 and GFP 215 showed a complex unfolding energy landscape of the latter. In a recent, high resolution force spectroscopy approach the equilibrium fluctuations of calmodulin, a calcium- dependent signal transducer, were studied. 59 It was shown that target peptides have stabilising function on folded calmodulin domains and that the folding kinetics of these domains is influenced by calcium ions. The AFM was also used to unzip the neck coiled coil domains of the motor protein kinesin-1. 216 As with nanoindentation experiments on microtubules and viruses and the force spectroscopy of unbinding experiments, also for unfolding measurements the comparison of experimental data with simulations have yielded additional insights into the exact molecular processes that take place. Examples include MD simulations on unfolding of specific titin domains, 217,218 titin kinase 213,219 and GFP. 220

2.6 Conclusion

In this review, we have shown the advances in the study of protein form and function using atomic force microscopy. The strength of the AFM in comparison to other biophysical techniques lies in its capability to image biological samples in their native environment and its versatility to do force spectroscopy. Hence, AFM can be considered as a worthy addition to the standard methods and techniques in protein studies as exempliﬁed by the experiments discussed in this review. With continuing development and advancement of the AFM, it is expected to become an indispensable tool in the study of single proteins and supramolecular protein assemblies and in elucidating the relationship between their form (structure) and function.



Chapter 3

Imaging at the (supra)molecular scale: from DNA, colloids to viruses

This chapter is a compilation of publications that I co-authored with other researchers. References to the actual papers are made at the respective sections.

Abstract

We demonstrate the capability of atomic force microscopy (AFM) to image particles and molecules which are too small to observe by light-based microscopes. From DNA- TFAM complexes, nanocolloidal-albumin particles to Norwalk viruses, we illustrate the imaging range (1 to 40 nm) and sub-nanometer resolution that AFM imaging can attain, in air as well as under liquid conditions. First, we investigated the compaction of DNA by TFAM and how it affects replication and transcription processes in mitochondria. We quantified the state of compaction by measuring the traceable contour length and end-to-end distance of the DNA using the AFM images as we vary the concentration of TFAM. We showed that the formation of DNA-TFAM complexes is a cooperative process and that at the physiological concentration of TFAM there is a very small window where both nearly-naked and fully-compacted DNA co-exist. This result suggests that the compaction of TFAM acts as a regulatory mechanism on the number of mitochondrial DNA available for replication and transcription. Secondly, we investigated the labeling of nanocolloid-albumin with radioactive isotopes 99mTc, 89Zr and fluorescent dye IRDye 800CW on the size of the nanocolloids and how it affects its uptake to the lymph nodes and cancerous tumuors in rabbits. We showed that the labeled nanocolloids have similar sizes as the unlabeled ones Introduction and the uptake is not affected for subsequent imaging through lymphoscintigraphy by PET/CT scans. Furthermore, we showed that the IRDye 800CW fluorescently- labeled nanocolloids have optimal kinetic and contrast features that could be a viable alternative to radioactive-labeling methods. Finally, we investigated the spontaneous assembly of the capsid protein VP1 into Norwalk virus-like particles and how it is regulated by its environment such as pH, ionic strength and protein concentration. The composition, shape and size of the particles were measured using mass spectrometry and atomic force microscopy. We showed that the native-sized Norwalk virus-like particles (VP180) are stable in a wide range of pH but their structural composition is sensitive to alkaline treatment. The broad ability of the capsid protein VP1 to form particles is exemplified by the observation that it assembles smaller T=1 VP160 at pH 9, intermediate VP180 particles at pH 8 and bigger T=3 VP1180 at pH < 7. By following the time-course of (dis)assembly of the particles by pH dilution from 6 to 9, we inferred that the VP1180 dissociates into smaller oligomers such as dimers and monomers and spontaneously- assemble to form the smaller VP160 particle. The VP180 particle seems to be a kinetically-trapped intermediate at pH 8 which is formed from partial dissociation of VP1180. These findings show the ability of VP1 to form particles under a wide range of conditions and its adaptability to withstand the fluctuating pH conditions during its oral-to-fecal route of infection.

3.1 Introduction

Since the development of atomic force microscopy by Binnig, Quate and Gerber in 1986, 17 the atomic force microscope (AFM) has become an indispensable imaging technique both in academic and industry settings, in a wide range of research fields such as solid state physics, materials and biological sciences. As a contact-based microscope, it does not suffer the diffraction limits of light-based microscopes thus it can reach subnanometer down to atomic resolutions. Moreover, the contact of the AFM tip and the sample and surface can be exploited to mechanically manipulate the sample to obtain its mechanical, structural and material properties. 52 In this chapter, we present three studies where we used the AFM to image molecules and supra-molecular assemblies, ranging from DNA, colloids to viruses. It en- compasses studies from factors affecting mitochondrial replication and transcription, developing new labels for imaging tumuors and cancer cells, to self-assembly of viral particles. Using AFM we were able to study the sizes, shapes, conformation and structure of these particles in relevant physiological conditions. The results obtained

 3 Imaging at the (supra)molecular scale by AFM are discussed in the light of other techniques we used such as biochemical assays, ﬂuorescence microscopy and mass spectrometry.

3.2 Methodology

DNA-TFAM complex preparation

The DNA used was a modified version of pBluescript plasmid (pBS-LSPH, 4348 bp) cut in ScaI position resulting in an unspecific linear DNA of 4348 bp. A stock solution of mouse Mitochondrial Transcription Factor A (TFAM) from a mouse with a concentration of 20 pmol/µL in (900 mM NaCl, 20 mM NaPO4, 10% Glycerol) buffer, was diluted in (5 mM Tris-HCl, 5 mM MgCl2, pH 7.6) buffer to yield a DNA/TFAM ratio of 10, 20 and 40 DNA bp per TFAM molecule. For the reaction, 1 µL of linear DNA (27 ng/µL) and 1 µL of TFAM (4, 2 or 1 pmol/µL) were incubated in 18 µL of the Tris-HCl buffer (described above) at room temperature for 10 minutes. The reaction mixture was diluted further to decrease the NaCl concentration to 3 mM prior to deposition on mica. A volume of 20-50 µL of the diluted mixture was deposited onto freshly-cleaved mica, allowed to adsorb for one minute, rinsed with milli-Q water and dried with a stream of N2 gas. The sample was directly imaged with an AFM (Bruker BioCatalyst System, Peak Force Tapping mode at 1.5 kHz) using a SiN cantilever (Olympus OMCL-RC800PSA, k = 0.8 N/m).

DNA contour length measurement

The contour length of the DNA was measured using a home-made LabVIEW program. The AFM images were processed as follows: flattening, binary thresholding, skeletonization and contour tracing. During flattening, each line of the image was fitted with a linear function and the corresponding linear fit is subtracted from each pixel value along that line. This is performed for all the vertical lines to correct for the vertical tilt in the image. This process is then repeated for the horizontal lines of the vertical tilt-corrected image to complete the 2D tilt correction of the image. The DNA-protein complexes were distinguished and differentiated from the background using binary thresholding. In this process, the pixel value histogram of the image was used to determine what intensity value separates the complexes from the background. Using this threshold, the gray scale image was transformed to a binary image, where white corresponds to the complexes (pixel value = 255) and black for the background (pixel value = 0). The DNA images were thinned up to about 3-5 pixels wide using a skeletonization routine in LabVIEW. The contour of the DNA was traced using eight 5 × 5 image masks corresponding to the eight nearest neighbors of each pixel. The

 Methodology

A B C

Figure 3.1 Steps in processing DNA images. The chosen region of interest containing one DNA molecule is: A) ﬂattened, then B) thresholded and skeletonized, and C) its contour is traced, from which the contour length and end-to-end distance are measured. The values of these parameters for this particular DNA are Lc=1375.3 nm and EED=424.3 nm as shown in the boxes below the pictures B and C.

mask which gives the highest weight determines the direction of the next pixel, and this process is repeated until the whole contour of the DNA is traced with 1-pixel wide resolution. The contour length of the DNA was corrected for the direction of the next pixel, i.e. 1 if two adjacent pixels are horizontal or vertical neighbors and 1.4 otherwise. This process is illustrated in figure 3.1. To quantify the fraction of the DNA that is compacted by TFAM, we define throughout this chapter the traceable contour length (Ltc) as the length of that part of the DNA molecule that does not participate in the compaction. In case the compaction makes the automated tracing and tracking difficult, this was performed manually.

Labeling of nanocolloidal albumin

The nanocolloids (from a 0.5-mg kit for the labeling of colloidal human serum albumin with 99mTc) was premodified with p-isothiocyanatobenzyldesferrioxamine B (Df-Bz- NCS [Macrocyclics]) and labeled with 89Zr(IBA Molecular) according to the proto- col. 221 The 99mTc-nanocolloidal albumin was prepared according to the manufacturer’s information manual. 222 In addition, native nanocolloidal albumin in polysor- batecitrate buffer, pH 6.5, diluted to a concentration of 0.10 mg/mL, was included in this analysis to determine whether the modification and subsequent labeling altered the particle size of the colloids. IRDye 800CW was coupled covalently to nanocolloidal albumin according to the following procedure. Tween-citrate buffer (TCB) was added to a Nanocoll kit vial, and 0.5 mg of nanocolloidal albumin was purified by sep- arating it from stannous chloride by size-exclusion chromatography (PD10 column,

 3 Imaging at the (supra)molecular scale

TCB pH 8.5 as eluent). Subsequently, 11.6 µg of IRDye 800CW-NHS ester (10 nmol in 20 µL DMSO) was added to 1 mL eluate containing the purified nanocolloidal albumin (0.25 mg, 3.7 nmol HSA building blocks, pH 8.5). This yields a 2.7 molar excess of the IRDye over HSA units and appeared to provide the product with optimal fluorescence yield. The mixture was incubated for 2 h at 35 ◦C. After 2 h, nonconju- gated IRDye 800CW was removed by size exclusion chromatography (PD10 column, TCB pH 6.5 as eluent). The void volume and the first 1.5 mL were discarded; the next 2 mL containing the nanocolloidal albumin-IRDye 800CW conjugate was used for further experiments. The AFM (Nanotec Electronica, Spain) was operated in jumping mode, and all experiments were performed in liquid. Rectangular cantilevers (Olympus OMCL- RC800PSA, k = 0.05 N/m) with a nominal tip radius of 20 nm were used to image the colloidal particles. Colloidal samples were deposited on a freshly cleaved mica substrate and incubated for 10 min before they were imaged by AFM. Height measurements on the immobilized particles were performed using a home-written program in LabVIEW (National Instruments, USA) to obtain the particle size distribution.

Norwalk virus-like particle preparation

The recombinant capsid protein VP1 was expressed in Spodoptera frugiperda (Sf9) cells using a baculovirus expression system and purified as described previously. 223 The Sf9 cells were harvested 5 to 7 days after baculovirus infection and then purified using centrifugation with a CsCl gradient of 1.362 g/cm3. The purified recombinant Norwalk Virus-Like Particles (rNVLPs) with a concentration of 400 µM VP1 were stored in water at 4 ◦C. The rNVLPs were buffer-exchanged into an aqueous ammonium acetate buffer (50-500 mM) at various pH values using an Amicon Ultra 0.5-mL centrifugal filter (Millipore, Billerica, MA) with a molecular mass cutoff of 10 kDa. The pH of the ammonium acetate solution was adjusted by using an aqueous solution of either ammonia or acetic acid. The samples were incubated in the respective buffer for at least 30 minutes. High resolution and tandem mass spectra were recorded on a modified Q-ToF 1 instrument (Waters) in positive ion mode. 224 To enhance the transmission of the large ions corresponding to rNVLPs, xenon, at a pressure of 2×10−2 mbar, was used in the collision cell. The voltages and pressures were also optimized for large non-covalent protein complexes. Briefly, the capillary and cone voltages were kept constant at 1450 and 165 V, respectively. The voltage before the collision cell was varied from 10 to 400 V but generally left at 50 V for the accumulation of native electrospray ionization mass spectrometry (ESI-MS) spectra. Ions were introduced into the source under an elevated pressure of 10 mbar. To generate intact gas phase ions from large

 Methodology protein complexes in solution, the source was maintained at 6.1 mbar, and voltages of 1400 and 164 V were applied to the capillary and sample cone, respectively. Xenon was used as the background gas in the trap and transfer ion guides at a flow rate of 4 mL/min. The voltages in the trap and transfer were 20 and 25 V, respectively. The gas in the ion mobility cell was nitrogen at a flow rate of 25 mL/min and a ramped wave height of 10-30 V with a velocity of 250 m/s. Ramped wave heights were shown previously to provide better results for protein complexes in the megadalton range. Collision cross sections (ccs) were determined from the measured drift times through calibration using proteins of known cross-sections. Denatured ESI-MS analysis of the VP1 monomers was performed on an LCT instrument (Waters). ESI tips were prepared in house from borosilicate glass tubes of 1.2-mm outer diameter and 0.68- mm inner diameter (World Precision Instruments, Sarasota, FL) by using a P-97 micropipette puller (Sutter Instruments, Novato, CA). The ESI tips were gold-coated using a Scancoat six Pirani 501 sputter coater (Edwards Laboratories, Milpitas, CA). The AFM experiments (Nanotec Electronica, Spain, jumping mode) were conducted in aqueous ammonium acetate solutions using rectangular cantilevers (Olym- pus OMCL-RC800PSA, k = 0.05 N/m). The cantilevers were calibrated as described previously 117 and were found to have a spring constant of 0.052 ± 0.004 N/m. The substrates used were hydrophobic-treated glass coverslips. They were cleaned by im- mersing in an 86% (v/v) ethanol solution saturated with potassium hydroxide for 14 h. After being rinsed thoroughly with deionized water and left to dry, the coverslips were rendered hydrophobic by incubating them under a saturated hexamethyldisilazane (Fluka, The Netherlands) vapor for 14 h. 169 The hydrophobic treatment of the substrate was made such that the virus particles are adhering to the surface but not desorbed or deformed due to surface-substrate interaction. This ensures that the structural as well as mechanical integrities of the particles are kept intact, while the jumping mode of operation of the AFM minimizes lateral shearing or dragging of the particles. A 100 µL droplet of sample containing 0.2-10 µM VP1 in 500 mM ammonium acetate buffer at various pH values was placed on the hydrophobic glass coverslip and imaged after a 15- to 30-min incubation period. The images were processed using the WSxM software 225 and the home-made LabVIEW program.

 3 Imaging at the (supra)molecular scale

3.3 Compaction of mitochondrial DNA by TFAM

This section is derived from: Nucleoid packaging blocks replication and transcription of mitochondrial DNA, (manuscript in preparation). 226

A mammalian cell contains about 103 to 104 copies of mitochondrial DNA (mtDNA), a circular molecule of 16,568 bp that encodes for 13 essential subunits of the respi- ratory chain. The mitochondrial genome is essential for normal cellular functions, and mtDNA mutations can cause mitochondrial diseases and have been implicated in human aging. 227,228 In vivo, mtDNA is not a naked molecule, but exists in a compact nucleoprotein complex called the nucleoid. In mammalian cells, mtDNA is fully coated by a protein denoted as mitochondrial transcription factor A (TFAM). In addition to its role as an mtDNA packaging factor, TFAM is also a component of the mtDNA transcription machinery. Besides TFAM, the minimal transcription machinery in mammalian cells consists of a nuclear-encoded RNA polymerase (POL- RMT) and mitochondrial transcription factor B2 (TFB2M). 229–232 In combination, these three factors can support transcription from a DNA fragment containing a mitochondrial promoter. TFAM binds sequence-specifically just upstream of the mitochondrial light- and heavy-strand promoters and induces a stable U-turn in DNA. How this stable kink in the DNA promote sequence specific transcription initiation remains to be established, but it may position TFAM to promote specific interactions with the mitochondrial transcription machinery. 233 The molecular composition of nucleoids have been carefully analyzed and, in addition to TFAM, a number of nucleoid-associated proteins has been identified, many of them with known roles in mtDNA transactions, e.g. POLγ, mtSSB, and TWINKLE. Even if these proteins play essential roles in nucleoid function, only TFAM has been defined as a structural component of the nucleoid. TFAM binding covers about 30 bp of DNA. Interestingly, in vivo estimates have determined the concentration of TFAM to one molecule per 10 to 20 bp of mtDNA. 234 TFAM binds in a cooperative manner forming stable protein patches on mtDNA. Binding partially unwinds the double- stranded structure, leading to a remarkable softening of the DNA and a compaction of the mtDNA molecule. 235 The compaction of DNA into a highly condensed structure is essential for the maintenance and the transmission of the mitochondrial genome. In support of this notion, disruption of the TFAM gene in mouse causes loss of mtDNA, whereas over- expression of TFAM leads to an increase in mtDNA copy number. However, the high

 Compaction of mitochondrial DNA by TFAM concentration of TFAM needed for DNA compaction will result in fully-coated DNA molecules, and might therefore affect the activity of DNA-binding molecular motors, such as the mitochondrial DNA and RNA polymerases, that have to progress on such busy tracks during replication and/or transcription. Here we will investigate using AFM the impact of TFAM on DNA packaging which in turn is believed to regulate replication and transcription. From the AFM images, the compaction of DNA with varying concentration of TFAM can be visualized and the degree of compaction can be quantified from the change in the traceable contour length and end-to-end distance of the DNA-TFAM complex. To do this, we incubated linearized DNA molecules with increasing concentrations of TFAM and recorded images of the nucleoprotein complexes formed. Figure 3.2 shows AFM images of these DNA-TFAM complexes deposited on mica and imaged in air. In vivo, there are about 1000 TFAM molecules per mtDNA, which roughly correspond to 15 ∼ 20 bp of mtDNA per TFAM molecule. The naked linear DNA molecules were readily observed as mostly fully-stretched out. At a lower DNA to TFAM ratio (40 bp/TFAM), we mostly observed one to three patches of bound TFAM molecules on the DNA, and on rare occasions a fully-coated but not compacted DNA (rightmost DNA-TFAM complex). As the TFAM concentration was increased further to levels similar to those in vivo (20 bp/TFAM), we observed different degrees of compaction, ranging from almost naked DNA to fully compacted nucleoprotein complexes. The co-existence of free DNA and highly compacted nucleoprotein complexes supports the idea that TFAM binds cooperatively to DNA. A further increase of TFAM concentrations to (10 bp/TFAM) caused nearly all DNA molecules to form compact nucleoprotein complexes. We measured the traceable contour lengths and end-to-end distances of the DNA molecules as shown in figure 3.2 B and C. TFAM packaging of DNA caused a dramatic decrease in the traceable contour length of the DNA, from ∼1.4 µm for naked DNA to less than 0.1 µm for completely covered DNA. This demonstrates the cooperative binding of TFAM and its ability to package DNA into compact, spherical nucleoprotein structures. This is also illustrated by the decrease of the DNA’s end-to-end distance from 0.6 µm to close to zero as shown in figure 3.2C. The results demonstrated that small changes in mtDNA:TFAM ratio had dramatic consequences for mtDNA packaging and at TFAM concentrations close to those observed in vivo, both compacted and nearly completely-naked DNA molecule could be observed and coexist at physiological TFAM concentrations. Thus, the variations in TFAM concentration can therefore epigenetically regulate the number of mtDNA molecules available for gene transcription and DNA replication.

 3 Imaging at the (supra)molecular scale

Figure 3.2 DNA compaction at diﬀerent TFAM concentrations. A) AFM images of TFAM-DNA complexes and distributions of the DNA B) traceable contour length (Ltc) and C) end-to-end distance (EED) at diﬀerent TFAM concentrations. The DNA has a length of 4348 bp ≈ 1.4 µm. The measurement of Lc and EED is described in the methods section. N≈150 particles for each concentration.

In the nucleus, there are only two copies available of each gene. Thus the cell has developed a highly-complicated system for gene regulation, with gene-specific transcription factors that stimulate the frequency of transcription. In contrast, the mitochondrial genome is present in 103 ∼ 104 copies in a somatic cell. This large number means that gene regulation may be regulated not only at the frequency of transcription of a specific mtDNA molecule, but also by the fraction of mtDNA molecules used for active transcription. There may be mechanisms that select only a subset of mtDNA molecules for replication, whereas others remain in a silent state. In fact, the number of TFAM molecules are directly proportional to the mtDNA copy number. 236 This indicates that it is essential for the cell to maintain the ratio of TFAM to mtDNA within a narrow range, since small variations may have dramatic functional effects for transcription and mtDNA replication. TFAM levels may be regulated in vivo by the Lon protease. 237 When cells were cultured in the presence of ethidium bromide, which cause mtDNA depletion, Lon

 Compaction of mitochondrial DNA by TFAM knock-down cells are unable to degrade TFAM, which in turn leads to a strong increase in the TFAM:mtDNA ratio. The change in TFAM to mtDNA ratio leads to an inhibition of mitochondrial transcription. This finding is in excellent agreement with results presented in this report and provides in vivo evidence for a role of TFAM as an epigenetic regulator. In fact, there is even evidence for regulation of TFAM levels by intracellular signaling pathways, since phosphorylation of TFAM by cAMP- dependent protein kinase in mitochondria impairs DNA binding and promotes Lon- dependent degradation. 238 This finding thereby provides a mechanism that allows for rapid changes in TFAM levels in response to cellular events. Consequently, this fine-tunes the fraction of mtDNA molecules available for active DNA replication and transcription.

Conclusion

We have shown using atomic force microscopy that TFAM compacts DNA in a cooperative manner. The results also show that at physiological concentrations, compacted and nearly-naked DNA co-exist. Thus, the variations in TFAM concentration can therefore regulate enzymatic activities where DNA unwinding is required, such as gene transcription and DNA replication. These processes are three orders of magnitude more eﬃcient in mitochondria than in the nucleus due to the higher number of DNA molecules available. The compaction of TFAM seems to be the regulation mechanism utilized by mitochondria to control and silence the other DNA molecules, thereby epigenetically regulating the fraction of mitochondrial genomes available for transcription and replication.

 3 Imaging at the (supra)molecular scale

3.4 Size of nanocolloidal-albumin for tracing tumuors

This section is derived from the publications: Journal of Nuclear Medicine 52 1580-1584, (2011). 221 European Journal of Nuclear Medicine and Molecular Imaging 39 1161-1168, (2012). 239

Lymphoscintigraphy or sentinel node (SN) mapping procedure is a diagnostic technique applied in a variety of tumuor and cancer types, including head and neck squamous cell carcinoma. 240,241 It identiﬁes the lymph drainage basin by using marked molecules that are taken up by tumuor cells such as albumin. Albumin is a major protein in blood plasma, which under stress condition such as in growing tumuor tissue, is taken up by cells as a source of amino acids and energy. Albumin are marked either with nuclide isotopes or dyes that can be detected by imaging techniques. The present standard procedure of SN mapping is preoperative lymphoscintigraphy using planar or SPECT (Single Photon Emission Computed Tomography) imaging, performed after peritumoral injections of 99mTc-labeled albumin colloid (generally called 99mTc-Nanocoll in Europe). Identifying sentinel nodes near the primary tumuor remains a problem in, for example, head and neck cancer because of the limited resolution of current lym- phoscintigraphic imaging when using 99mTc-nanocolloidal albumin. We demonstrate the development and evaluation of a nanocolloidal albumin-based tracer speciﬁcally dedicated for high-resolution PET (Positron Emission Tomography) detection. The coupling of 89Zr to nanocolloidal albumin was made via the bi-functional chelate p-

24 nm A 1 B

1 2

2 3

0 nm Figure 3.3 Nanocoll size distribution as assessed by AFM imaging. A) AFM image of Nanocoll particles (non-labeled) with height proﬁles of three selected particles. B) Height distribution of the non-labeled, 99mTc-labeled, 89Zr-labeled and IRDye 800CW ﬂuorescence-labeled Nanocolloidal albumin, yielding average particle sizes of 13.8 ± 0.8 nm, 14.9 ± 0.5 nm, 12.7 ± 0.6 nm and 14.6 ± 0.4 nm. N=100∼200 particles for each sample.

 Size of nanocolloidal-albumin for tracing tumuors isothiocyanatobenzyldesferrioxamine B. Quality control tests, including particle size measurements, and in vivo biodistribution and imaging experiments in a rabbit lymphogenic metastasis model were performed. Coupling of 89Zr to nanocolloidal albumin appeared to be efficient, resulting in a stable product with a radiochemical purity greater than 95%, without affecting the particle size as shown in figure 3.3. The measurement of the size of the particles can be used to determine the effect of labeling on the nanocolloids with different isotopes, such as aggregation or a sub- stantial increase in the particle size after labeling which could affect the uptake of the nanocolloids. PET showed distinguished uptake of 89Zr-nanocolloidal albumin in the sentinel nodes, with visualization of lymphatic vessels, and with a biodistribution comparable to 99mTc-nanocolloidal albumin as illustrated in figure 3.4A and B. This study shows that 89Zr-nanocolloidal albumin is a promising tracer for sentinel node detection by PET. 221 At present, the only approved fluorescent tracer for clinical near-infrared fluorescence guided sentinel node detection is indocyanine green (ICG), but the use of this tracer is limited due to its poor retention in the SN resulting in the detection of higher tier nodes. We present the development and characterization of a next-generation fluorescent tracer, the nanocolloidal albumin-IRDye 800CW that has optimal properties for clinical SN detection. The fluorescent dye IRDye 800CW was covalently coupled to colloidal human serum albumin (HSA) particles present in the labeling kit Nanocoll in a manner compliant with current Good Manufacturing Practice.

A B C d e

f g

Figure 3.4 A) Static planar lymphoscintigraphy image of a rabbit with VX2 auricular carcinoma after the injection of 99mTc-nanocolloidal albumin. B) PET image after the injection of 89Zr- nanocolloidal albumin. The parotid (big arrow) and caudal mandibular (small arrow) lymph nodes are visible on both images, but the the contralateral injection site (?) is also visible in B. C) In vivo evaluation of fluorescently-labeled nanocolloidal albumin-IRDye 800CW on a rabbit with VX2 auricular carcinoma. NIR fluorescence image of fluorescence-labeled lymph nodes obtained 5 min and 24 h after peritumoral injection of: d,e) nanocolloidal albumin-IRDye 800CW and f,g) ICG/HSA (1=parotid, 2=caudal mandibular lymph nodes, a=angle of mandible, T=tumuor).

 3 Imaging at the (supra)molecular scale

Labeling of premodified nanocolloidal albumin with 89Zr proved reproducible, with an overall labeling yield of 70%-75%. After PD10 purification, radiochemical purity always exceeded 95% and remained stable on storage for at least 24 h. The covalent conjugation of the fluorescent dye IRDye 800CW to nanocolloidal albumin is very simple and reproducible, and leads to about 14 IRDye 800CW groups per nanocolloidal albumin particle, yielding a strong fluorescence signal intensity. In addition, the size of the nanocolloidal albumin particles was not altered after the respective conjugation of the different labels and is comparable with that of the standard 99mTc-Nanocoll as shown in figure 3.3. Characterization of nanocolloidal albumin-IRDye 800CW included determination of conjugation efficiency, purity, stability and particle size. Quantum yield was determined in serum and compared to that of ICG. For in vivo evaluation a lymphogenic metastatic tumuor model in rabbits was used. Fluorescence imaging was performed directly after peritumoral injection of nanocolloidal albumin-IRDye 800CW or the reference ICG/HSA (i.e. ICG mixed with HSA), and was repeated after 24 h, after which fluorescent lymph nodes were excised. Conjugation of IRDye 800CW to nanocolloidal albumin was always about 50% efficient and resulted in a stable and pure product. The quantum yield of nanocolloidal albumin-IRDye 800CW was similar to that of ICG. In vivo evaluation revealed noninvasive detection of the SN within 5 min of injection of either nanocolloidal albumin-IRDye 800CWor ICG/HSA. No decrease in the fluorescence signal from SN was observed 24 h after injection of the nanocolloidal albumin-IRDye 800CW, while a strong decrease or complete disappear- ance of the fluorescence signal was seen 24 h after injection of ICG/HSA as illustrated in figure 3.4 d-g. This shows that nanocolloidal albumin-IRDye 800CW is a promising fluorescent tracer with optimal kinetic features for SN detection. 239

Conclusion

We have described the production and preclinical evaluation of two alternative labeling techniques of nanocolloidal albumin: (1) the 89Zr-nanocolloidal albumin as a PET tracer specifically dedicated to lymphoscintigraphy by PET/CT and (2) IRDye 800CW-conjugated nanocolloidal albumin as a promising next generation fluorescent tracer with characteristics that would allow intraoperative SN detection even after a longer interval of time. The AFM proved to be a useful method to evaluate the size effect of labeling and characterize the very small (10 to 20 nm) unlabeled/labeled particles. We have shown that labeling the nanocolloids with different isotopes or a fluorescent tracer did not significantly change the size of the particles and no aggregation was observed. Furthermore, no differences were observed between 89Zr and the IRDye 800CW-conjugated nanocolloidal albumins as compared to the stan-

 Size of nanocolloidal-albumin for tracing tumuors dard 99mTc-nanocolloidal albumin regarding radiochemical purity, SN detection by imaging, and uptake in the regional lymph nodes. These results justify the evaluation of 89Zr-nanocolloidal albumin in a clinical setting as an additional technique side-by-side the standard 99mTc-nanocolloidal albumin. In addition, the evaluation of IRDye 800CW conjugated nanocolloidal albumins opens a ﬂuorescence-guided detection method as an alternative to conventional radioactive-guided methods which shows optimal imaging, uptake and kinetic features and allows longer time interval for SN detection.

 3 Imaging at the (supra)molecular scale

3.5 The pH-regulated assembly of Norwalk virus-like particles

This section is adapted from the publication: Molecular & Cellular Proteomics 9 1742-1751, (2010). 122

The assembly of viral capsid, in which viral proteins self-assemble into complexes of well defined architecture, is a fascinating biological process. The study of viral structures and assembly processes have been the subject of many excellent structural biology studies in the past but questions still remain regarding the intricate mechanisms that underlie viral structure, stability, and assembly. Although detailed structural information on the fully assembled capsid exists, less information is available on potential capsid (dis)assembly intermediates, largely because of the inherent heterogeneity and complexity of the disassembly pathways. We used native mass spectrometry (MS) and atomic force microscopy (AFM) to investigate the (dis)assembly of the Norwalk virus-like particles as a function of solution pH, ionic strength, and the concentration of the basic structural protein VP1. 52,122 The norovirus represents an important human pathogen, accounting for most cases of non-bacterial gastroenteritis. 242 It is the most predominant pathogen within the family Caliciviridae, which also includes Sapovirus, Vesivirus, and Lagovirus. 243 The prototypical strain of the genus Norovirus is the Norwalk virus. It is a small and non-enveloped virus. Its 7.7-kb of single-stranded RNA genome contains three open reading frames, encoding for the major capsid protein (VP1), the minor capsid protein (VP2), and a non-structural polyprotein. 244 The mature Norwalk virus capsids (T=3) are composed of 90 VP1 homodimers 120,121 and possibly a few copies of VP2 that are thought to stabilize the icosahedral structure as well as affect the expression of VP1. 245,246 Because of the lack of suitable animal models or in vitro cell culture systems, structural studies so far have been largely focused on recombinant Norwalk virus-like particles (rNVLPs), which are spontaneously assembled during the expression of recombinant VP1 and VP2 in insect cells. These empty noninfectious particles have been demonstrated to be morphologically and antigenically similar to the genuine virion. 223 The rNVLPs have been studied extensively using X-ray crystallography and electron microscopy (EM), which have provided a detailed image of the intact capsid, revealing the T=3 icosahedral organization. 121,245 The VP1 monomer structure is principally composed of two domains, a shell (S) domain and a protruding (P) domain as illustrated in figure 3.5. In the intact capsid, the S domain forms a contiguous

 The pH-regulated assembly of Norwalk virus-like particles

A) B)

Figure 3.5 Electron microscopy reconstructions of rNVLP. A) Surface view showing the capsid shell and protrusions. B) Cross-section view showing the VP1 monomer and its shell (S) and protruding (P1 and P2) domains. Modiﬁed from Bertolotti-Ciarlet et al. 245 and Prasad et al., 121 reproduced with permission (copyright © 2002, American Society for Microbiology) and © 1999, American Association for the Advancement of Science).

protein shell with a diameter of ∼30 nm, whereas the P domain forms prominent protrusions, which give the rNVLPs a diameter of ∼40 nm. A remarkable feature of the rNVLPs is that only a single protein (VP1) is responsible for directing capsid assembly and host interactions. The rNVLPs thus represent a simple model to study the assembly of icosahedral viruses. Although the requirements for capsid assembly have been investigated previously, there is little information regarding intermediates along the (dis)assembly pathway. 245 Obtaining such information can be quite diﬃcult because of the inherent heterogeneity of capsid assembly. An emerging technique for interrogating such heterogeneous protein assemblies is native electrospray ionization mass spectrometry (ESI-MS). 53 In addition, we used AFM imaging in complementary to the ESI-MS technique to investigate the sizes of the assembled/disassembled particles and intermediates during buﬀer exchange and pH dilution.

Assembly/disassembly monitored by mass spectrometry

The stability of rNVLPs over a range of solution conditions has been probed using EM and various spectroscopic techniques. 245 From this work, it was discovered that although the rNVLPs are stable under acidic conditions they are quite sensitive to even slight alkaline treatment. Under acidic pH, ions corresponding exclusively to the intact capsid were observed in the mass spectrum. Even at pH 2.5, intact rNVLPs dominated the spectrum, although some small VP1 oligomers, predominantly VP1 monomers and dimers, were also detected. Conversely, upon alkaline treatment, extensive capsid dissociation was observed. Representative mass spectra of the rNVLPs in a 250 mM ammonium acetate buﬀer over a pH range of 6-9 are given in ﬁgure 3.6A.

 3 Imaging at the (supra)molecular scale

The ion intensities for all species were summed for each solution pH and are represented in the bar graph keeping all other variables such as VP1 concentration and ionic strength constant. Figure 3.6A shows that at pH 8, extensive capsid dissociation was observed as VP1 dimers were quite abundant in the spectrum. Several (dis)assembly intermediates were also detected, with the VP160 species being the most populated of the VP1 oligomers. Interestingly, these smaller capsids coincide to those reported in a previous EM study. 247 These smaller particles possessed a diameter of around 25 nm and were speculated to be T=1 particles made up of 60 copies of VP1. Thus, it appears that at this pH the intact capsid can disassemble, initially forming smaller VP1 oligomers, predominantly dimers. These small VP1 oligomers can then interact to reform higher order oligomers, predominantly VP160 at pH 8. It has been suggested

A B

pH 7 C

pH 8 D

pH 9

Figure 3.6 Stability of rNVLPs over a range of pH and ionic strength monitored by native ESI- MS. A) Mass spectra obtained from aqueous solutions of the rNVLPs (30 µM VP1) at different pH values in 250 mM ammonium acetate buffer . The different fractions of VP1 oligomers are summarized in the bar graphs. B), C), D) The effect of ionic strength of ammonium acetate in the range of 50-500 mM on the assembly of VP1 oligomers at a pH 7, 8 and 9, respectively.

 The pH-regulated assembly of Norwalk virus-like particles that in solution the VP1 dimers may exist in two distinct conformations and that the N terminus of VP1 acts as a switch to distinguish between these two dimer conformations guiding T=3 icosahedral assembly. 121,248 It seems that there are subtle pH-induced conformational changes within the VP1 dimers, affecting the switching mechanism and leading to the formation of alternate capsids with different symmetry and stoichiometry. A further increase to pH=9 led to the complete loss of the full VP1180 capsid and the formation of more mid-size particles such as the VP180 and VP160 species and the rest are in the form of low-order oligomers. To investigate the effect of ionic strength on the assembly of these VP1 oligomers, ESI-MS analysis was performed after dilution of the intact rNVLPs into various ammonium acetate buffers with a concentration range of 50-500 mM and a pH range of 7-9. The ion intensities for each VP1 oligomer were summed overall charge states at each solution composition and are summarized as bar graphs in figures 3.6B, C and D. At physiological pH 7, the intact VP1180 capsid are robust at different ionic strengths as shown in figure 3.6B. However under alkaline conditions, e.g. at pH 8, the intact VP1180 capsid are stable only at ionic strength of 100 mM as shown in figure 3.6C. Below 100 mM, few of the VP1180 capsids disassembled and VP180 are formed, while above 100 mM the formation of the smaller VP160 becomes more favorable. It is interesting to note that at even higher ionic strengths, a decrease in the VP1 dimers is concomitant to the increase in VP160. Thus, it appears that above a certain ionic strength the assembly VP1180 capsids becomes unstable and eventually disassemble to smaller oligomers, most probably VP1 dimers and then reassemble into the VP160 species. These trends are even more evident at pH 9 as shown in figure 3.6C. At this pH, there was no intact VP1180 at any of the ionic strengths investigated. As was observed at pH 8, at higher ionic strengths, there was an increasing propensity to form the VP160 and VP180 particles, with a concomitant decrease in dimer (VP12) abundance. In fact, below a certain ionic strength threshold (100 mM), there were no higher order oligomers present as almost exclusively VP1 dimers were detected in the mass spectrum while at higher ionic strength, the assembly of VP160 is more favorable than VP180. Another important factor in the formation of protein assemblies is the concentration of their protein building blocks. To investigate the effect of VP1 concentration on the assembly of these novel VP1 oligomers, serial dilutions of a VP1 stock solution (160 µM) in a 250 mM ammonium acetate buffer at pH 9 were made and analyzed by ESI-MS. It was found that at low concentrations of VP1 the large oligomers were very low in abundance, and VP1 dimer was the dominant species in the spectrum. With increasing VP1 concentration, VP160 and VP180 became more prevalent. Inter- estingly, the ratio of VP160/VP180 was also highly dependent on VP1 concentration,

 3 Imaging at the (supra)molecular scale

that is the VP180 species becomes more abundant than the VP160 oligomer. The assembly of these higher order VP1 oligomers was also found to be a reversible process. These experiments demonstrate the amazing ability of the VP1 to reversibly and rapidly form higher/lower order structures depending upon the solution conditions. It has been recently suggested from molecular dynamics simulations of capsid assembly that reversible steps are essential to capsid formation as they diminish the eﬀect of kinetic traps. 249 This appears to be the case for the rNVLPs and the VP1 oligomers described here and may explain why Norwalk virus is such a robust and environmentally persistent virion.

Assembled particles imaged by atomic force microscopy

We further investigated the (dis)assembly of rNVLPs at acidic/alkaline pH in physiological buﬀer conditions by imaging the particles using AFM. Figure 3.7 shows sample AFM images of the rNVLPs. These particles were deposited on hydrophobic cover slips in buﬀer conditions. Figure 3.7A and B show high density of rNVLPs deposited on the surface. The concentration of the particles was gradually diluted

AB C

D E F

1 2

Figure 3.7 Sample AFM images of rNVLPs. Images of high particle density of rNVLPs deposited on hydrophobic glass cover slip at pH 6: A) 1 × 1 µm, B) 0.5 × 0.5 µm. Images of low particle density: C) 1 × 1 µm, D) 0.5 × 0.5 µm. E) 3D image and F) height proﬁles of the particles in E.

 The pH-regulated assembly of Norwalk virus-like particles such that each particle is isolated enough to make the height profile measurements easy as shown in figures 3.7C-F. From the height profile, we determined the diameter of the particles. To scrutinize the disassembly and assembly process of the different particles seen in the MS analysis, the AFM images were collected under the same buffer conditions. From the MS data, it is clear that the formation of the particles requires high ionic strength thus, VP1 was diluted in 500 mM ammonium acetate buffer. Figure 3.8A shows the height distributions of the particles formed when VP1 proteins are stored in pH 6 and pH 9. The particles at pH 6 were found to be spherical and possess an average diameter of 39.0 ± 0.3 nm, which is in good agreement with the expected 121 value of 38 nm for VP1180 obtained from x-ray crystallographic studies. On the other hand, at pH 9 no VP1180 particles were detected and the formed particles have an average size of 22.6 ± 0.2 nm, which corresponds well to the smaller 23-nm VP160 particle obtained from electron microscopy studies. 247 The particles were also spherical, suggesting that this smaller VP1 oligomer is a capsid of different stoichiometry and symmetry. Combining information regarding stoichiometry obtained from MS with the structural data afforded by AFM provides strong evidence that the VP160 oligomer is a spherical capsid with T=1 icosahedral symmetry. To investigate the reversibiliy of the assembly/disassembly, buffer exchange experiments were also performed from pH 6 to pH 9 and vice versa as shown in figures 3.8B. It clearly shows that the (dis)assemly of the VP1180 and VP160 upon pH change is reversible. The assembly of the larger VP1180 particle at pH 6 is complete and stable in this buffer condition as illustrated by the narrow distributon centered at 38 nm. Although most of the particles at pH 9 possess diameters of 23 nm, there were some larger species detected which likely corresponds to VP180 with diameter of about 32 nm. However, because of their low abundance, it is difficult to make any definitive conclusions regarding the morphology of the VP180 species. At this concentration of VP1 (10 µM) and at this solution composition (high ionic strength and high pH), very little VP180 was detected by MS experiments as well. A higher VP1 concentration should lead to the formation of more VP180. However, it is difficult to perform the AFM analysis at such high VP1 concentrations because the glass plate becomes over- crowded with particles, precluding the height measurements of individual particles. We investigated further the disassembly process by diluting the pH from 6 to 9 and observed the time course of its disassembly by imaging and recording the sizes of the particles formed. Figures 3.8C shows that the disassembly of the VP1180 is relatively fast as shown by the presence of smaller particles right after dilution, but the assembly process of the VP180 and VP160 seems very slow. After 4 hrs, VP160 and smaller oligomers (< 12 nm) were formed, but there is also a significant percentage of VP1180

 3 Imaging at the (supra)molecular scale

Figure 3.8 Assembly and disassembly of rNVLPs. A) Particles formed at pH 6 and pH 9. B) Re- versible (dis)assembly of VP1180 and VP160 by buﬀer exchange. C) Time lapse imaging of the disassembly of VP1180 to VP160 by pH dilution.

particles. After 24 and 72 hrs, the VP1180 particles have largely diminished but the fraction of VP160 and VP180 remained the same. Instead, smaller oligomers (10 to 15 nm) become the dominant species. These results seem to indicate that the VP1180 particles disassembled first to smaller oligomers such as dimers and monomers and then assembles into smaller VP160 particles. Unfortunately, the sizes of the dimers and monomers (< 5 nm) can not be imaged and measured accurately due to the roughness of the glass surface. The results seem to indicate that the VP160 particles take a long time to reach equilibrium conditions at alkaline conditions. This rather slow and incomplete assembly of VP160 may be attributed to the low concentration of monomers or dimers and surface effects from the glass cover slip. The condition at pH 8 presents a saddle point in the energy landscape of (dis)assembly between the T=1 VP160 and T=3 VP1180 particles. We speculate that VP180 particles are kinetically-trapped intermediates at pH 8 and are partially-dissociated remnants of VP1180. The apparent instability of the smaller particles formed at higher pH indicate that these particles are likely formed at the intestinal lumen and likely do not contain any genome inside. 247 Although these particles are energetically- and structurally-viable, the size and structure of the wild type VP1180 and the presence of the genome produce a more stable virion. Further studies on genome-filled capsids will be of interest to determine the effect of the genome on the structural stability of the particles.

Conclusion

Using native mass spectrometry and atomic force microscopy analyses, we have shown that the recombinant Norwalk virus-like particles are very stable under acidic conditions but sensitive upon alkaline treatment. At elevated pH, intermediate structures were observed of which the VP180 and VP160 were found to be the most abundant species and formed as spherical particles. The VP160 species exhibit the T=1 icosa-

 Summary

hedral symmetry but the exact structure of VP180 is still not known. In addition, we have shown that the assembly and disassembly of these particles are completely reversible upon buffer exchange or pH dilution. The results suggest that the disassembly process undergoes first the dissociation of the VP1180 particles upon alkaline treatment into smaller oligomers, predominantly dimers before it reassembled into the VP160 species. This suggests that the intermediate VP180 species is a quasi-stable structure that formed during the dissociation process of the VP1180particle. It needs to knock-off ten VP1 dimers or dissociate completely to reassemble the more stable VP160. The results illustrate the ability of the capsid protein VP1 to adapt to its environment and exhibit its capability to form smaller or bigger particles to obtain stable capsid structures at different pH conditions.

3.6 Summary

This chapter compiles the work I did with other scientists and showcases the capabilities of the AFM as a versatile machine to investigate biophysical problems by imaging biomolecular structures and features which cannot be addressed using light- based microscopy. With the DNA-TFAM complexes, we were able to visualize the 1-nm DNA strands and their 2D conformation on the mica surface. The visual information and quantification of compaction using AFM provided a strong verification of the results of the bio-chemical assays of our study. The measurement of the size of the labeled nanocolloidal-albumin using AFM also provided a straight-forward and direct way to verify that the 10-20 nm size of the nanocolloids are unaffected by labeling and that there are no aggregation effects after using the novel labeling iso- tope 89Zr and fluorescent dye IRDye 800CW instead of the standard 99mTc. These labeling techniques open up two additional techniques such as PET scan with 89Zr and fluorescence microsocopy with IRDye 800CW for imaging tumuors and cancer cells. Finally, we demonstrate the capability of the AFM to image viruses in physiological buffer conditions. The images provide information on the size and structure of the assembled 15-40 nm Norwalk virus-like particles, including their assembly and disassembly behavior. These results provide insights on the development of synthetic nanoparticles for drug/gene delivery systems and other applications. All in all, these studies show that AFM imaging can be performed on different kinds of biological materials in dried as well as physiological conditions.

 Chapter 4

Microscopy and force spectroscopy of viruses

Published as Soft Matter 6 5273-5285, (2010). 18

Abstract

The recent developments in virus research and the application of functional viral particles in nanotechnology and medicine rely on sophisticated imaging and manipulation techniques at nanometer resolution in liquid, air and vacuum. Atomic force microscopy (AFM) is a tool that combines these requirements and imaging of viruses dates back to the early days of AFM. AFM has a comparable resolution to electron microscopy, but has the advantage that it can be used to image in fluid. This implies that dynamic processes in physiologically relevant environments can be studied and one of the first examples of AFM imaging of viruses was the real-time observation of viral infection of living cells. Improvement of the imaging techniques led the way to image fragile viruses in non-destructive ways, to visualise viral capsomeres in liquid and to image genome uncoating. In addition to imaging, AFM is also used as a tool to manipulate viral nanoparticles. Translation, rotation and nano-dissection of viruses are possible and by performing nanoindentation experiments the viral material properties can be examined. Next to providing the Young’s modulus and breaking force of viral shells, such experiments have also elucidated the impact of the genome on the overall viral mechanical properties and the effect of capsid structural modifications. The combination of modelling and AFM experiments finally, yields a deeper insight into their structure, function and behaviour. Here we review the early AFM experiments on viruses, the achievements made since then and the recent advances in imaging and manipulation. Introduction

4.1 Introduction

Within a few years after the publication of Binnig, Quate and Gerber describing the atomic force microscope 17 in 1986, the broad potential of imaging viruses and virus- like particles by atomic force microscopy (AFM) was recognized. On one hand viral particles were used as a calibration tool: viruses generally exhibit a regular and identical structure, from one particle to the next, therefore they were used to determine AFM tip geometry 119 and as a control specimen to test imaging parameters. 250 On the other hand, they were imaged to study their structure as well as fundamental biological processes. Initial experiments showed hints of DNA release from lysed bacteriophages 141 and the presence of tail fibers. 251,252 Furthermore viral infection of single cells was studied. 134 In the following years the imaging techniques were refined and manipulation of single viruses by AFM became possible as well. This manipulation was either directed to move, nano-dissect or destroy the particles, 253,254 or to obtain insight into their material properties 83,169,254 The process of improvement and broadening of the techniques still continues and this review covers the major achievements in imaging and manipulation of viruses by AFM up to this day. Viral capsids are generally symmetric, closed structures which are formed by self- assembly of several copies of identical or a limited amount of different proteins. For example, the minute virus of mice (MVM) 255 and the cowpea chlorotic mottle virus (CCMV) 256 capsids are composed of only one type of protein while phage λ capsids 257,258 and Φ29 proheads 259 have two and four different constituent proteins, respectively. Such viral structures are often elucidated by x-ray crystallography or electron microscopy techniques, imaging the viruses in atomic detail. These images corroborate the hypothesis of Crick and Watson in 1956 that capsids are composed of identical subunits that are packed together in a regular manner. 260 Most of the spherical viruses exhibit icosahedral symmetry in which the constituent proteins are arranged into pentameric and hexameric units called capsomeres. Caspar and Klug’s theory of quasi-equivalence describes this special architecture and classifies the viral capsids using the so called triangulation number T. 261 A visualisation of how to determine the T number of a capsid can for instance be found in ref. 262 According to this classification, MVM, CCMV, phage λ and herpes simplex virus 1 (HSV1) have a triangulation number of T=1, 3, 7 and 16 respectively. Some viruses can also assemble in two or more forms such as the hepatitis B virus (HBV) which self-assembles in either a T=3 or T=4 configuration. Aside from sphere-like viruses a large class of viruses has a tube-like appearance of which tobacco mosaic virus (TMV) is a well known example. 263 Others can be a hybrid of different classes such as bacteriophage Φ29 which possesses a special prolate structure composed of icosahedral end caps and a cylindrical equatorial region. 264

 4 Microscopy and force spectroscopy of viruses

Viruses are intensely studied to elucidate their assembly, structure and infection process either from the point of view of fundamental research or to cure viral diseases. Over the last ∼20 years they are also increasingly being studied for their use as functional nanoparticles in medicine and nanotechnology. Applications range from nanocontainers for catalysis and drug delivery via virus based therapies and diagnostic tools, to building blocks in materials science, (electro)chemistry and biophysics. 265–268 For these studies a reliable imaging technique is needed that can operate in air, liquid and vacuum at nanometer resolution. The atomic force microscope is exactly such a technique and in the next section we will brieﬂy discuss the AFM set-up.

4.2 Atomic Force Microscopy

AFM is unique in its possibility to study protein samples in air or liquid, with nanometer resolution. It allows imaging of dynamic systems as well as conducting force measurements, giving it wide-spread applicability in biological, physical and material sciences. Figure 4.1 shows a schematic of AFM (also sometimes referred to as Scanning Force Microscopy, SFM) showing its major components. AFM imaging can be performed in various modes, contact mode and tapping mode being the ones

Figure 4.1 Schematic of an AFM set-up. This simple schematic shows the basic elements of an atomic force microscope. The piezo moves the sample in the x- and y-direction in a scanning motion for imaging and in the z-direction to respond to surface topography changes or to apply a force. Cantilever deﬂection is recorded by the quadrant photodiode via a change in reﬂection of the laser light. The radius of curvature of the tip (situated at the end of the cantilever) is one of the determinants for the maximum achievable resolution. Imaging can be performed in air as well as in liquid. Other geometries than the one depicted here are common, for instance where the x and y and/or z piezo scanners are attached to the cantilever instead of the sample.

 Imaging of viruses and genome release most commonly known. Contact mode AFM is the traditionally used mode, but is generally rather invasive and tends to destroy fragile samples. Tapping mode AFM was developed a few years later and has the advantage that it can be operated in a more gentle way to minimise sample deformation. With the advent of tapping mode techniques in liquid, 269,270 a wide range of possibilities were opened to study biological samples in a physiologically relevant environment and with nanometer resolution, but without destroying the sample. However, even with tapping mode AFM it is important to set the imaging parameters correctly, otherwise fragile samples – like viruses – will be damaged. 271 Variations on tapping mode AFM are available and viruses have for instance also been imaged by jumping and magnetic AC mode AFM. 94,272,273 In addition to imaging, force measurements on biological samples can be conducted using AFM. In such experiments the AFM probe pulls upon or pushes into a sample and its material properties can be deduced from the force-distance curves. Whereas several papers have reviewed AFM measurements on biological substrates in general, 87,88,131,274–278 this review focuses in detail on AFM experiments with viruses.

4.3 Imaging of viruses and genome release

4.3.1 AFM imaging of viruses

In the early 1990s a couple of groups started imaging viruses, with albeit different goals in mind. The first publications on AFM visualisation of viruses appeared in 1992 however, they were not so much oriented to study virological questions but rather to the use of viruses as a calibration and reference tool. The cylindrical viruses tobacco mosaic virus (TMV), tobacco etch virus (TEV) and bacteriophage M13 were adhered to mica and imaged in air to deduce the shape of AFM tips. 119 TMV was furthermore used as a resolution criterion for sample preparation and to test the imaging parameters of AFM. 250 Whereas the measured dimensions of TMV turned out to be comparable to cryo-transmission electron microscopy data, the authors noted that the tip geometry and imaging force of 10 nN in air are serious limitations to high resolution imaging of biological material. This was confirmed by a study that investigated the influence of the tip radius on particle compressibility. 279 While imaging TMV with blunt tips (radius of curvature Rc ∼ 57 nm) at a probe force of ∼16 nN gave a correct particle height, imaging at a comparable force with sharp tips (Rc ∼ 6 nm) resulted in compression or disintegration of the particles. Besides cylindrical viruses, icosahedral viruses with a tail, like bacteriophage T4, were also imaged. These studies were more oriented towards investigating the structure and function of viruses than those discussed above. T4 phages, adhered onto a silicon

 4 Microscopy and force spectroscopy of viruses

Figure 4.2 AFM images of BMV crystals. (A) Crystal of T=3 BMV particles, capsomeres can be distinguished on the capsid surfaces. (B) T=1 particles obtained after CaCl2 treatment of T=3. BMV capsids can also be crystallised and surface features are observed on the individual particles. Reprinted with permission from ref. 135 (Copyright © 2001, Society for General Microbiology).

substrate, were imaged in air. Intact viruses as well as lysed viruses were observed, with the latter showing DNA strands lying next to it. 141 Apart from imaging the head and the tail, it was also shown that it is possible to image the tail fibers of phage T4. 251,252,280 By functionalising mica surfaces with amine groups, bacteriophage fd was imaged in liquid. 281 Real time attachment of these particles was followed over 40 minutes, showing a steady increase in phage adhesion. However, it was observed that the attachment of the particles to the surface was not as good as for phages that had been dried first and then re-immersed in liquid. This is likely related to the use of the rather invasive contact mode for imaging. While using less invasive imaging techniques is one option to increase resolution of particles in liquid without involving a temporary drying step; improving the attachment of the particles onto the substrate is another way to obtain this goal. Clearly this only works for samples that are not easily destroyed. By using glass cover slips that were functionalised with a photoactivatable crosslinker, a considerable increase in imaging resolution was obtained. Polyheads of bacteriophage T4 were covalently bound to this functionalised glass by irradiation with ultraviolet light. 282 This assured a firm attachment of the particles, needed for high resolution imaging. While imaging under buffer conditions, the gp23 hexamers of the T4 polyheads could be neatly resolved, a major step forward in the high resolution imaging of viral structures. In order to decrease the interaction forces between tip and sample, dynamic force microscopy was used to image TMV particles in air. 283,284 With the advent of tapping mode AFM in liquid 269,270 it became possible to image viral particles under physiological conditions without applying high lateral forces that are common in contact

 Imaging of viruses and genome release mode imaging. However, contact mode is still used for imaging of viruses, especially when the particles are properly immobilised. Contact mode imaging has for instance been used for the real-time imaging of crystal growth of satellite tobacco mosaic virus (STMV). 285,286 In this study it was shown that crystal growth occurs through nu- cleation in two and three dimensions leading to the formation of so-called “stacks” of 10∼20 monolayers. After 1995, having established the possibility of imaging viruses by AFM, a considerable number of publications appeared on this topic. AFM imaging of brome mosaic virus (BMV), a T=3 plant virus, revealed the capsomeres on its surface (Fig- 135,287 ure 4.2A). Treatment with CaCl2 induced the formation of T=1 BMV particles and even on these smaller particles it was possible to distinguish the capsomeres, as shown in figure 4.2B. Furthermore, it was possible to image the capsomeres on other viruses like for instance turnip yellow mosaic virus 288 and cauliflower mosaic virus. 135 AFM imaging also revealed specific structures of various viral particles. For instance, it unveiled a specific segmented morphology of the tail of the filamentous closterovirus beet yellows virus, which was not seen on the control samples potato virus X and TMV. 289 Aside from topographic imaging, the use of the phase information in tapping mode AFM can further unravel sample properties. Phase contrast imaging of phage Φ29 revealed information on the local elasticity of structural components like the virus collar and hollow tail knob, indicating a different structural integrity as compared to the capsid. 127 Cryo-electron microscopy (cryo-EM) and AFM were used for high resolution imaging of mimi virus, the largest known virus. It is an icosahedral double stranded DNA virus with a diameter of ∼750 nm 290,291 and the imaging showed a clear picture of a starfish-shaped feature present on a unique five-fold vertex of each viral particle. 292 It is suggested that the star-fish shaped feature has the function of a seal, preventing premature DNA delivery. Aside from imaging the outside surface of viral capsids, the inside features of broken capsids were also imaged. In particular, broken as well as intact bacteriophages ΦKZ were imaged with AFM in air. 293 These phages were not directly manipulated by AFM, instead the particles were disrupted by washing the adsorbed viruses with water and subsequently an uranyl acetate solution. AFM images of partially disintegrated phage ΦKZ showed the presence of a high object, apparently coming from the capsid. This could be an internal protein body, of which hints were also found in cryo- EM reconstructions of this phage. 294 Disassembly of potato virus X was induced by complexing with movement protein (virally encoded proteins that allow for cell-to- cell transport of plant viral genomes) 295,296 and subsequent AFM imaging. 297 The authors suggest that the binding of movement protein triggers destabilisation of the particle, leading to disassembly of the particle while being imaged by AFM.

 4 Microscopy and force spectroscopy of viruses

AFM was used to study the dynamic binding, unbinding and reorganisation of genetically engineered cowpea mosaic virus on nanoscale chemical templates in liquid. 298 Analogies to atomic-scale epitaxial systems as well as to colloidal systems in conﬁned geometries were observed. In addition to following the binding of single viral particles to substrates, binding of molecules to single viruses has also been studied. In this study, the binding of receptor molecules to human rhinovirus 2 (HRV2) was followed under physiological conditions. 299 HRV2 was physisorbed onto mica in the presence of divalent cations and receptor molecules were subsequently added. One, two or three receptor molecules, with a diameter of ∼8 nm and a height of ∼1.5 nm, were bound per virus. Furthermore it was shown that the dynamics of binding could be followed by imaging the same viruses with a 10-minute interval. Attachment as well as detachment events were recorded. Imaging of the bare viruses, without receptors, revealed the protrusions at the 3- and 5-fold symmetry axes. Surface symmetry features of the T=1 icosahedral shell of another virus, the parvovirus minute virus of mice, were also imaged. 185 The viruses were immobilised on hydrophobic glass cover slips 169 and from visual inspection of the images, it was possible to discern on which of the three symmetry axes the particles adhered to the surface. AFM imaging of herpes simplex virus 1 (HSV1) in air was performed on intact virions and stripped particles. 300 In particular, images were obtained from (i) enveloped particles, (ii) particles without the envelope, but with attached tegument and (iii) bare capsids. Images of the envelope revealed hints of the presence of membrane proteins. Images of the capsid, albeit distorted as a result of surface attachment and drying, revealed the pentons and hexons. Additionally it has been shown that imaging of HSV1 under physiological conditions yields a comparable resolution on undeformed particles. 138,187 Triangular faces of the icosahedral HSV1 capsids as well as the pentons and hexons were imaged in buﬀer. Furthermore, it was shown that it is possible to image capsids which lack the pentameric vertices. 138 Figure 4.3 shows a HSV1 capsid after extraction of the pentons with 2.0 M GuHCl, 301,302 visualising the impressive stability of these capsid structures despite the presence of regular defects.

4.3.2 Studying cell-independent genome release by AFM

Genome release from viral capsids has been studied by various means 180,305,306 and to reduce the number of free parameters, systems without cells are often used, i.e. release with only the capsid and genome present. As mentioned previously, one of the ﬁrst publications on AFM imaging of viruses was on lysed bacteriophage T4 capsids. 141 The authors suggested that lysis occurs as a result of osmotic shock or surface attachment. Lying next to the lysed particles, bundles are observed that are likely DNA strands coming out of the phage. The average height of the lysed particles is

 Imaging of viruses and genome release

Figure 4.3 AFM imaging in buﬀer of HSV1 capsids. The pentameric vertices of this capsid are extracted with 2.0 M Guanidine hydrochloride (GuHCl). (a) Top view image of a penton-less capsid lying on its 3-fold symmetry axis. The arrows point to three holes at the position of the missing pentons. (b) Enlargement of the central part of the same particle with additional contrast enhancement. (c) Same image as above, with additional marking of the (missing) capsomeres. The vertices where the pentons are missing are denoted with 5 (in red), the peripentonal hexons with P, the edge hexons with E and the central hexons with C (using the notation of refs. 303,304). (d) Cross-sections along the two-fold symmetry axes. Curve I shows the height proﬁle over the entire capsid and curve II (lower inset) a zoom in on the top part of this particle. The cross- sections clearly reveal the P and E hexons as well as the holes where pentons are missing. The lateral distance between the hexons corresponds to about 16 to 17 nm, which is consistent with data obtained from cryo-EM reconstructions. 304 Reprinted with permission from ref. 138 (Proceedings of the National Academy of Sciences U.S.A., 2009).

approximately half the height of the intact particles, corroborating the conclusion that the genome indeed came out from these particles. Using various methods known from literature to extrude the genome from viruses, uncoated TMV and turnip yellow mosaic virus were observed by AFM imaging. Stripping of TMV particles to expose their RNA after treatment with mild alkali solutions, urea or dimethylsulfoxide resulted in structures that were imaged by AFM in air. 307 Other harsh treatments, including drying and rehydrating steps or heating, were used in order to image RNA extrusion out of turnip yellow mosaic virus capsids. 308 Furthermore, drying of bacteriophage Φ29 and minute virus of mice resulted in genome expulsion which was conﬁrmed by AFM imaging. 309 Herpes simplex virions that were treated with 0.5% sodium dodecyl sulfate were shown to spill their DNA onto the surface. 300 HSV1 capsid interactions with the nuclear pore complex (NPC) were imaged in an approach to visualize genome delivery. 310 Imaging was performed after spreading and subsequently air drying the nuclear envelope of HSV1 infected Xenopus laevis oocytes on functionalised glass cover slips. The cytoplasmic as well as the nucleoplasmic sides of the nuclear envelope were imaged and thick rod-like structures were observed. The authors suggested that the DNA transfers the NPC as a condensed rod-like structure, however the possibility that the results reﬂect an artefact of air drying the sample,

 4 Microscopy and force spectroscopy of viruses

Figure 4.4 HRV2 genome uncoating. (A,B) Partially released RNA (arrow in A) was shown to be digested enzymatically after injection of RNase and an incubation time of 10 min (arrow in B). (C,D) Partially as well as completely ejected RNA (respectively lower and upper arrows in (C)) are both shown to be digested by RNase (arrows in D). (E) Imaging over 40 min of partially and completely ejected RNA in the absence of RNase. The steady position of the RNA molecules in this control experiment shows that mere imaging does not displace, remove or destroy the RNA molecules. Reprinted with permission from ref. 142 (Copyright © 2004, American Society for Microbiology).

cannot be ruled out. Other viruses that were imaged together with their genome include cauliflower mosaic virus and moloney murine leukaemia virus. 135 Human rhinovirus 2 (HRV2) is taken up in endosomes during the infection of host cells and the genome is released from the capsids by a lowering of the pH inside the endosome. 311 Whereas cryo-EM images revealed the opening of twelve vertices in empty capsids, 312 presumably to facilitate genome uncoating, RNA had not been visualised in the process of uncoating until AFM imaging was performed. 142 A droplet of HRV2 in buffer solution was deposited onto a mica surface and after a brief incubation the pH was lowered from 7.6 to 4.1 and raised again to its original value after 2 hours. This treatment induced RNA expulsion in a near-physiological way and enabled AFM imaging of the extruded genome. Figure 4.4 shows how the ex- pulsed RNA is digested after injection of RNase into the sample, corroborating the conclusion that the observed filaments are indeed RNA molecules. The majority of

 Imaging virus-cell interactions the imaged RNA molecules were probably still partially inside their capsid because the genome length outside the capsid was 330 nm. Some molecules however seemed to be completely expelled and measured maximally 1 µm.

4.4 Imaging virus-cell interactions

AFM experiments of the infection of monkey kidney cells by pox viruses (in particular vaccinia virus) were first reported in 1992. 134 These experiments were performed under physiological conditions over time ranges of up to 45 hours. Single cells were held in place by a micropipette and were imaged prior to, and after infection. Approximately 2–3 hours after infection significant protrusions occurred on the cell membrane, which seemed however too small to be viruses. These protrusions were interpreted as reactions of the cell to the production of viral particles. 313 After ∼20 hours microvilli occurred, showing resemblance to those observed by EM imaging of vaccinia virus release. 314 This process, occurring on time scales of ∼10 min, 134,139 was therefore associated with the exocytosis of viruses. However, the reported softening of the cell membrane, 134 minutes after infection, was not reproduced in later experiments by the same group. Infection of fibroblasts by the retrovirus murine leukemia virus (MLV) was studied with chemically fixed 315,316 and living cells. 140 Fixed cells without the membrane or with the membrane (preserved by osmium tetroxide) were imaged in ethanol and showed the viruses in late stages of morphogenesis or already in the course of budding. 315 Cells infected with a mutant that produces viral particles with an apparently defective maturation protein were shown to produce tube-like structures on the cell surface. This was interpreted as the incapacity of mutant virions to undergo wild type morphogenesis and budding. Figure 4.5 shows images of viral budding from living cells. 140 These experiments reveal that there seem to be two kinetically distinct mechanisms by which the virions bud. A part of the virions (∼25%) bud within 25 minutes, whereas the majority of the particles (∼75%) bud in a process taking more than 45 minutes. Interestingly, the growth rate of the particles in these two groups are similar, but whereas the particles in the “fast” group are released quickly, in the “slow” group there seems to be an arrested phase at the end of growth. So the assembly rate for both groups seems similar. Hence, this arrested phase can be interpreted as an energetic barrier for fission, which is supported by modelling approaches of fission events. 317 The authors argue that the observed difference in kinetics of budding could reflect a difference in biochemical activity of the cell. Fi- nally they observe that virions generally do not bud from the same site on the cell, challenging the hypothesis that budding occurs at distinct positions along the cell

 4 Microscopy and force spectroscopy of viruses

Figure 4.5 Imaging of living fibroblasts infected by MLV. (A) Tapping mode AFM image in liquid of an uninfected living fibroblast. Filamentous structures observed through the membrane are likely cytoskeletal components. Scale bar: 5 µm. (B) Image of a MLV infected living fibroblast. Viral particles in various stages of budding can be observed. Scale bar: 3 µm. (C) Complete budding process followed by AFM imaging. At t=0 min an emerging bud can be observed which grows to its maximal height ∼20 min later. Just before t=48 min the virion is finally released. The top panels show the AFM height images and the lower panels a cross sectional height profile. Scale bar: 0.5 µm. Reprinted with permission from ref. 140 (Copyright © 2008, The Biophysical Society Published by Elsevier Inc.).

surface. In a follow-up paper the role and dynamics of the actin cytoskeleton during budding of MLV and human immunodeficiency virus (HIV), both retroviruses, was studied. 125 The actin cytoskeleton, which forms a cross-linked network beneath the cell membrane, 34,318,319 is shown to be actively remodelled by the budding virions into aster-like structures, as was visualised by AFM torsion mode imaging. Mutant viruses that do not bind actin bud an order of magnitude slower, underlining the role of the cellular cytoskeleton in retroviral assembly and budding. Previous to the live cell imaging of HIV budding, this process was already described for virions budding from lymphocytes that were fixed by glutaraldehyde before AFM imaging. 320 Whereas some of the infected cells showed little to no HIV particles on their surface, others were covered by at least a hundred virions. This was interpreted as a dependence of viral budding on the cell cycle or physiology. Other experiments with chemically fixed cells were performed with the semliki forest virus and Theiler’s

 Manipulation of viruses in air murine encephalomyelitis virus, showing a cell surface smoothening post infection. 321 In a cell membrane mimicking approach, virus-membrane attachment was studied by AFM. By preparing receptor containing membranes which were adhered on mica sheets, a dense packing of human rhinovirus 2 could be obtained. 322 The images, recorded in buffer, show a quasi-crystalline arrangement of the viruses on top of the lipid bilayers. Next to interactions of eukaryotic viruses with their host cells, also prokaryotic virus (bacteriophage) infection was studied by AFM. Notably, interactions of various phages and their specific host cells were studied in air 323,324 and with air-dried samples which were reimmersed in liquid. 324 Infection of Escherichia coli by the filamentous phage M13 was shown to result in a smoothening of the cell surface. 324 Force-distance curves performed on these bacteria showed a decrease in stiffness and Young’s modulus after infection with M13. A similar decrease was observed for non- infected bacteria treated with 100 mM EDTA. As treatment with this concentration of EDTA was reported to decrease the amount of lipopolysaccharide (LPS) on the bacterial surface by roughly 50%, 325 it is suggested that these AFM results show that infection with M13 results in the damage and partial release of the bacterial LPS, as was previously proposed. 326

4.5 Manipulation of viruses in air

Viral capsids are increasingly being used in patterning and self-assembly approaches as nanocontainers for genome and drug delivery, and as nanoreactors for catalysis in constrained environments, 265–268 It is increasingly being recognised that an exact control of the physical properties of these nanoparticles is essential for creating optimal functionality. 327 One way to study and obtain a better insight on the physical properties of capsids is by using AFM. The atomic force microscope is well suited to determine for instance, the mechanical properties of viruses, as targeted manipulation of the particles can be directly performed after imaging. Several such studies have been conducted in air and in liquid. Whereas in surface science applications it can be important to know the mechanical properties of dried samples, in biology and medicine one is generally interested in studies performed in physiologically relevant environments, i.e. in liquid. Therefore this section is divided in experiments conducted in air (5.1) and in liquid (5.2). In a study to test the application of gold particles as a height and imaging standard in atomic force microscopy, these particles were co-imaged with tobacco mosaic virus (TMV) in air. 279 This also allowed studying the compressibility of TMV by tips of various radii. It was estimated that there is measurable compression of the

 4 Microscopy and force spectroscopy of viruses

Figure 4.6 Nano-dissection, rotation and translation of graphite adsorbed TMV particles by AFM manipulation in air. (a) Original TMV particle. (b) Dissection of the virus into two smaller particles. (c) Rotation of the top particle. (d) Translation of the particle. (e–f) Straightening of the particle, making it parallel to the lower particle. Total image width ∼1.3 µm, particle height ∼20 nm. Reprinted with permission from ref. 254 (Copyright © 1997, The Biophysical Society Published by Elsevier Inc.).

18 nm diameter TMV tubes above a tip normal pressure of ∼1 MPa. At a pressure of ∼40 MPa it seemed that there was irreversible damage to the viruses. Further- more, mica absorbed particles of TMV were shown to be damaged after repetitive scanning in contact mode. 253 This scanning resulted in the removal of the top half of these tube-like viruses. Nano-dissection of TMV was performed by repetitive line- scanning, showing a dependence of the cutting depth on the number of scans. Apart from nano-dissection on mica surfaces, 253 nano-dissection, translation, rotation and bending of graphite adsorbed TMV by AFM were reported as shown in figure 4.6. 254 This allowed deducing mechanical properties of TMV by studying the lateral friction between sample and substrate. Under the assumption that the shear stress between 328 graphite and TMV is comparable to reported shear stresses between MoS2/MoO3 and between Cd arachidate films, 329 a Young’s modulus of ∼1 GPa was found for TMV in air. Another cylindrical viral structure, the tail tube of bacteriophage T4, could be imaged repeatedly for a tip loading force of <4 nN. 330 However, when this force was increased to >5 nN the tubes gradually disappeared over the time course of multiple scans. In a proof of principle experiment in air, it was shown that normal force measurements on viruses could be used to gain insights on viral mechanical properties. 331 In this study, adenovirus was indented and a force indentation curve was fitted by the Hertz model for two indenting elastic spheres. Force-distance curves were also used to study mechanical properties of TMV particles. This yielded a rough estimate of

 Manipulation of viruses in liquid

5 GPa for its Young’s modulus, where a viral particle was probed that was supposed to be partially suspended with one end on the mica and the other on a second TMV particle. 332 However, there remain quite a few uncertainties in this approach. A more accurate approach seems to be that of nanoindentation measurements on surface adsorbed TMV particles where the force-distance curves are compared to ﬁnite element models of indentation. This approach yielded a Young’s modulus of 1.1±0.2 GPa for TMV in air. 333

4.6 Manipulation of viruses in liquid

Physical virology is a rapidly expanding field where virological questions are approached by physics-based methods. 180,181,334 Frequently-asked questions concern: the amount of force that is needed to package DNA with a molecular packaging motor, the pressure-if any-which exists inside viral capsids, the material properties of viruses, etc. Studies on prokaryotic viruses (bacteriophages) for instance, have revealed that these nanometer-sized capsids confine genomes which are several microns long. In this confinement, the genome (e.g. dsDNA in phage λ or Φ29) is packed to near-crystalline densities creating internal pressures of tens of atmospheres inside the capsid. 335–337 This implies that these capsids must have the mechanical strength to withstand the internal pressure due to genome packaging, in addition to the external stress received from the environment. This ensures successful safeguarding of its genomic content throughout the infection process. Much as bacteriophages were central in the development of molecular biology and genetic engineering techniques, it is suggested that pressurised viruses (most of the dsDNA bacteriophages are presumably pressurised) will play key roles in the development of the emerging field of physical virology. 181 Here we will show that indeed key insights on the mechanical properties of viruses are obtained by studying pressurised viruses, but nonetheless also on viruses which presumably lack such an internal pressure. Nanoindentation experiments using AFM have led the way to investigate the mechanical properties of viral capsids under physiological conditions, i.e. in liquid. 83,169 The relation between the force exerted by the AFM tip and the amount of indentation of the capsid surface provides information about its spring constant, Young’s modulus and the maximum force it can withstand prior to irreversible deformation or rupture. Several studies have compared the mechanical properties of viral capsids and long before the development of the atomic force microscope, these properties were already investigated by various bulk methods. For instance by using osmotic shock, whereby bacteriophages were incubated in highly concentrated salt solutions and subsequently rapidly diluted. 338 This treatment ruptured the capsids and was

 4 Microscopy and force spectroscopy of viruses

Figure 4.7 FZ curves recorded during a nanoindentation experiment in buffer solution. These force- indentation curves depict HSV1 capsid response upon deformation. The system was calibrated by bending the cantilever on the glass slide, yielding the reference curve FZ glass. The indentation response of the capsid is depicted by FZ forward, and the FZ backward curve shows the retraction of the cantilever. The indentation of this capsid shows a clear drop of the force, within the linear indentation part. The hysteresis between forward and backward curve shows the irreversibility of indentation, resulting from capsid failure. FZ 2 and FZ 3 show a strongly increased flexibility of the particle for additional pushing cycles, corroborating the indications that the particle is broken. These large scale disruptions appear permanent since waiting for several hours did not result in reversal to the initial mechanical properties of the capsids. Multiple breaking points are sometimes recorded during indentation; the presented particle in this figure shows two clear breaking events. The piezo displacement is denoted by Z displacement (see also fig 4.1). Indentation of the particle is the difference between the FZ glass curve and the FZ curve on the capsid. The FZ glass curve was shifted along the x-axis to have a coinciding contact point with the FZ virus curves. Adapted with permission from ref. 138 (Proceedings of the National Academy of Sciences U.S.A., 2009).

accompanied by release of the genome. However, with the advent of AFM techniques in liquid it became possible to study the material properties of single particles under buffer conditions. Using AFM, viral capsids can now be simultaneously imaged and physically probed under physiological conditions and their dynamic mechanical properties can be quantitatively measured. The first nanoindentation experiments on buffer-immersed viruses were reported in 2004, for empty capsids of phage Φ29. 169 Whereas this was the first report of a systematic study of the material properties of viruses in a physiologically relevant environment, it was not the first publication on AFM manipulation of viruses in liquid. The first example was the manipulation of adenovirus in liquid and was published a few years earlier. 339 In this study, it was shown that it is possible to move around a single adenovirus on a silicon substrate in water by AFM. However, no material properties of adenovirus could be deduced from these experiments, instead they mainly revealed insight into the adhesion force between virus and surface. On the contrary, material properties were probed during the nanoindentation experiment on phage Φ29. Capsids were immobilised, while being immersed in buffer, on a hy-

 Manipulation of viruses in liquid drophobic glass slide. 169 The atomic force microscope was operated in jumping mode, whereby shear forces on the sample are minimised by vertically retracting the tip from the sample before it is laterally displaced. 94 After taking high-resolution AFM images of the capsid, the tip was positioned at its centre and the sample was approached by the tip. Force-distance (FZ) curves were recorded by measuring the cantilever deflection (which corresponds to the tensile force applied by the tip) as a function of the vertical position of the piezo to which the sample is mounted. Figure 4.7 shows an example of such FZ curves (albeit on a different capsid, but the general tendency is similar). It was found that the Young’s modulus of the Φ29 procapsid is equivalent to that of hard plastic (∼1.8 GPa) and that the capsid exhibits a linear elastic behaviour at small indentations. However, beyond indentations of ∼12 nm, corresponding to 30% of its diameter and an applied force of ∼2.8 nN, it suffers irreparable damage and breaks. This study gives an indication of the mechanical perturbations that Φ29 procapsids can withstand when it is being packaged with its genome and building up its high internal pressure.

4.6.1 Eﬀect of genome packaging

Experiments on phage Φ29 show that an internal pressure of up to 60 atm can be present inside phage capsids. 335 Even though viruses which do not use a packaging motor probably have a smaller or negligible pressure inside their genome filled capsids, these results triggered questions on how the presence of the genome modifies the mechanical properties of capsids. Nanoindentation experiments were carried out on empty and full capsids of various prokaryotic and eukaryotic viruses: CCMV, 183 MVM, 185,340 phage λ 186 and HSV1. 138,187 Results on CCMV, MVM and phage λ reported stiffening and/or strengthening (increase in breaking force) of the capsid when it was filled with the wild type (wt) genome as compared to an empty capsid. Interestingly, no difference in mechanical properties was observed for empty and DNA filled nuclear HSV1 capsids. 138 In this section we will briefly discuss these experiments. Capsid stiffening upon genome encapsidation of CCMV and MVM is proposed to be the result of genome-capsid interactions. Coarse-grained RNA models combined with Monte Carlo simulations indicate a strong interaction between genome and capsid for CCMV as a result of RNA interacting with positively charged N terminal residues inside the viral capsid. 341 These interactions likely create a RNA/protein composite causing an effective increase in the capsid wall thickness, explaining the measured increased stiffness for genome filled CCMV. 183 Carrasco et al. reported an isotropic Young’s modulus for empty MVM capsids, but a symmetry axis dependent stiffening of genome-filled capsids. 185 A 3%, 40% and 140% increase at the 5-, 3-

 4 Microscopy and force spectroscopy of viruses and 2-fold symmetry axes was observed, respectively. They ascribed the increased stiffness to the interaction of the genome with the capsid wall, bearing similarities to the interactions described for CCMV. This hypothesis was affirmed by invoking protein engineering whereby capsid proteins were selectively mutated at positions that interact with the DNA. 340 They found that DNA patches/stretches interacting specifically at the capsid concavity at the 2-fold axes act as a molecular buttress thereby increasing the stiffness along this orientation. However, the 5-fold axis is vir- tually not affected by this structural reinforcement, likely due to its probable location as the entry/exit of the genome. A different mechanism to explain capsid stiffening upon genome encapsidation was proposed for phage λ. Ivanovska et al. measured the mechanical properties of phage λ with variable amounts of packaged genome lengths. 186 Remarkably, almost identical mechanical properties were found for partially-filled (78%– and 94%–filled with DNA) and empty capsids. However, a completely filled capsid (i.e. with wt genome length) becomes twice as stiff and stronger by 60% as compared to an empty one. This effect is attributed to the osmotic pressure generated by the DNA-hydrating water molecules, which increases exponentially as the packaged genome reaches a critical packing density. The same theory also explains the contrasting result in HSV1, where no difference in mechanical properties was observed between DNA filled and empty capsids. 138Calculating the density of the dsDNA inside phage λ and HSV1 yields a density inside HSV1 of approximately 60% compared to the density inside wt phage λ. As phage λ mutants with a genome length of 78% behave like empty particles, 186 a similar behaviour is expected and observed for empty and DNA filled HSV1 capsids. 138 Thus, the DNA-hydrating water molecules do only marginally affect the mechanical properties of DNA-filled HSV1 capsids, effectively yielding a similar mechanical behaviour of empty and DNA-filled capsids. The results described in ref. 138 furthermore explain the seemingly contradicting observation of a difference in mechanical behaviour of DNA filled and empty viral capsids reported by Liashkovich et al. 187 In this study DNA filled, tegumented capsids were treated with 1 M GuHCl, which removes the DNA from the capsids. 342 However, next to the DNA, tegument proteins are also extracted while treating viral capsids with GuHCl. 138 So the observed decrease in stiffness after treatment with 1 M GuHCl is probably not due to the presence or absence of DNA, but due to the reduction of tegument proteins present on the capsids. This nicely reconciles the observations on the mechanical properties of HSV1 capsids reported in ref. 187 and ref., 138 showing that the presence of DNA does not have an effect on the mechanical properties of HSV1 capsids.

 Manipulation of viruses in liquid

4.6.2 Changes in material properties during maturation

Retroviruses mature after budding from the host cell and undergo dramatic morpho- logical changes during this process. To understand the underlying physical mechanism of retrovirus maturation, indentation experiments have been performed on capsids before and after maturation. Measurements on moloney murine leukemia virus (MLV) 191 and human immunodeficiency virus (HIV) 47 reported a general trend of softening of the capsid during maturation. That is, immature capsids are stiffer than mature ones. Interestingly, not only the shell’s stiffness changes during maturation, also its thickness. It decreases from ∼20 nm to ∼4 nm, whereas the outer radius stays constant. So even though a mature MLV capsid is two times softer than an immature capsid, its Young’s modulus of elasticity is four times higher – because of its largely decreased shell thickness – indicating a stronger protein-protein interaction per unit length. The particle softening upon maturation seems to act as a stiffness switch for retroviruses to facilitate viral entry into cells. 47,191 Assembly of HSV1 capsids is initiated in the nucleus by the formation of a spherical procapsid around scaffolding proteins. 304,343–345 As a next step the procapsids mature while undergoing massive conformational changes to form a stable, mature capsid; a process that exhibits analogies to maturation of bacteriophages. 346,347 During this process the scaffold is expelled, DNA is packaged and the virus attains an icosahedral shape. 348,349 The maturation process however, seems to result in some faulty byproducts. Next to the genome containing C capsids also scaffold-containing B capsids and empty A capsids are formed, which possess the same angularised morphology as C capsids. 343–345,350 Nanoindentation experiments on HSV1 capsids (Fig. 4.7) have shown that B capsids break at a lower force than C capsids. 138 From these experiments and experiments on capsids without pentons (Fig. 4.3) it was concluded that scaffold expulsion and DNA packaging trigger a stabilisation process to make the capsids sturdier and prepare them for nuclear egress.

4.6.3 Eﬀect of structural modiﬁcations in the capsid

To investigate which structural components of capsids are essential to their dynamic mechanical properties, nanoindentation experiments were performed on capsids with structural modiﬁcations. Michel et al. conducted experiments on mutated CCMV capsids. 183 A single-point mutation on the capsid protein makes it more stable so it does not dissociate at conditions of pH 7.5 at high ionic strength, conditions which will dissociate the capsid and RNA of the wild type particles. 351 The experiments resulted in a higher Young’s modulus for the mutant (∼190 MPa) compared to the wild type (∼140 MPa) indicating that the mutation causes a stiﬀening as well as

 4 Microscopy and force spectroscopy of viruses strengthening of the capsid. This result shows that, to a certain extent, the mechanical properties of the capsid can be manipulated or engineered to have the desired physical characteristics. These results are highly interesting in the light of possible applications in nanotechnology and medicine where viral capsids can be used as nanocontainers for drug or gene transport and various other purposes. 267,268,352 The mechanical properties of CCMV were furthermore shown to be influenced by changes in pH. 50 Albeit not a directed structural modification of the capsid proteins, changes in pH have a striking influence on capsid stability, as shown by these experiments. Whereas CCMV capsids rupture upon nanoindentation at pH 5, this is not the case at pH 6. At the latter pH an elastic deformation of the capsids is recorded until a complete flattening of the particle. This visualises the obvious effect of how tuning the pH of the buffer solution can drastically change the protein-protein interactions inside capsids. Interestingly, both the T=3 and T=4 shells of hepatitis B virus have the same material properties. 184 This shows that differences in the structure of the capsid, do not necessarily lead to a change in material properties. Studies on HIV, an enveloped retrovirus, showed that truncation of the envelope proteins drastically decreased the stiffness of immature particles. 47 On the other hand, the stiffness of mature particles was almost unchanged after removal of the envelope proteins. These results suggest that, contrary to what is expected, the relatively high stiffness value of the immature particles is not principally caused by having a thicker shell than the mature particles. Rather, the presence of the cytoplasmic tail is one of the main contributors to its material properties. By specifically mutating MVM capsid proteins, the spring constant along the 2- and 3-fold axis of DNA- filled particles was significantly reduced. 340 These results support the view that a DNA-mediated mechanical reinforcement of the MVM virus capsid is caused by non- covalent interactions between a specific amino acid of the capsid’s inner wall and the packaged DNA. 185 Recent nanoindentation experiments on HSV1 where the icosahedral vertices of the capsids (pentons) were extracted by treatment with 2.0 M GuHCl (Fig. 4.3), show their impact on flexibility and rupture force of icosahedral capsids. 138 The results indicate the importance of the vertex capsomeres in maintaining the icosa- hedron’s structural stability. However, while the spring constant of the penton-less particles clearly decreases with respect to the penton-containing particles, the overall capsid structure is retained. This shows the amazing robustness of HSV1 particles, which, with only hexons present, remain intact.

 Combining experiments and modelling

4.7 Combining experiments and modelling

Whereas a wealth of theoretical and modelling studies on the mechanical properties of viral capsids have emerged (see e.g. refs. 189,262,353–359 and a recent review 188) we will here give a few examples of the combination of AFM nanoindentation experiments and modelling. Finite element modelling was used to extract the Young’s modulus from nanoindentation experiments. Initially this was performed with simple capsid models, which lack specific surface features. These capsid models, for instance of phage Φ29 and MVM, were indented by a point loading force and the deformation was analysed. 169,185 To model AFM nanoindentation with greater accuracy, the point loading force was replaced by a realistically shaped model of an AFM tip. Such a finite element model was used to analyse experiments on CCMV. 183 Finally, to increase the level of accuracy even further, instead of homogeneously thick capsid shell models, finite element models that incorporate the heterogeneous geometry of a viral capsid have been constructed. Using such capsid models, the difference in material properties of the native and swollen form of CCMV capsids is predicted with a non-linear continuum elasticity model. 190 This approach is likely to be used in the future for precise analysis of nanoindentation experiments on viral capsids. The results of nanoindentation experiments on TMV nanotubes were compared with finite element analyses as well as a Hertz model. 192 Both models, when combined with the experimental data, yielded Young’s moduli of ∼1.0 GPa. This value, obtained from TMV manipulation in liquid, is comparable to results reported on dried TMV substrates. 254,333 To compare the mechanical properties of CCMV at pH 5 and pH 6 a dimensionless control parameter, the Föppl-von Kármán (FvK) number was used. 50 The FvK number, the ratio of stretching and bending energies of a shell, has been used for theoretical descriptions of the morphology and deformation of icosahedral viruses. 189,262,353,357 Above a critical FvK value shells possess a more faceted shape, whereas they appear more spherical below this value. 353 The experimental indentation results of CCMV at pH 5 showed a discontinuity in the deformation response, corresponding to finite element simulations of thin shells with a FvK value of ∼900. 50 At pH 6 however, CCMV capsids show deformation behaviour which com- pares to particles with a FvK number of ∼100. The pH-tunable mechanical response of CCMV illustrates the application of the FvK number to describe experimental capsid deformation measurements. Aside from top-down modelling approaches like finite element analyses, bottom-up modelling approaches are also being used to describe virus nanoindentation experiments. As a result of the increase in computing power, molecular dynamics (MD) simulations of complete viruses are becoming feasible nowadays. MD simulations on small viruses have been performed in all-atom approaches 360 and for bigger viruses,

 4 Microscopy and force spectroscopy of viruses for which the computing power is not yet suﬃcient to perform all-atom simulations, coarse-grained approaches have been applied. 361 In particular, in an approach to model AFM nanoindentation experiments using MD simulations, an all-atom simulation of the deformation of the T=3 shell of the southern bean mosaic virus was performed. 362 However, in these simulations a very small indenter was used, making it diﬃcult to compare the results to AFM deformation experiments. In contrast, a realistically sized tip was used for coarse-grained MD simulations of the deformation of hepatitis B virus (HBV). 182 This work was motivated by AFM nanoindentation measurements on the T=3 and T=4 capsids of HBV, yielding, for both shells, the same Young’s modulus and critical force at which there is irreversible deformation. 184 The MD studies were focused on the irreversible deformation of the T=4 HBV shells and showed that the permanent deformation of the shells at high indentation forces is not the result of global rearrangements of capsid proteins, but rather of local shifting and bending. 182

4.8 Conclusion

In this review we have shown that AFM imaging of viruses has undergone significant improvements since the first images were published in 1992. Furthermore, with the advent and steady development of AFM manipulation techniques, an increased insight into the structure and material properties of viruses has been gained. Still there remain many open questions. For instance, whereas the viral maturation process has been studied successfully by nanoindentation experiments of retroviruses, the mechanics of this process for other eukaryotic viruses and bacteriophages remains elu- sive. The development of increasingly sophisticated modelling approaches 182,190,362 combined with AFM nanoindentation experiments promises to yield deeper insights into the molecular mechanisms of capsid (dis)assembly and viral deformation under stress. The recent progress in AFM high speed imaging 75,363 opens up possibilities for dynamic measurements on virus-cell interactions at a high temporal resolution as well as for studies of capsid assembly. The successes already obtained in viral studies by AFM together with the current developments promise a continuous research effort in the future, increasing our insight into the fundamental structure and function of viruses and their application in nanotechnology and medicine.



Chapter 5

Prestress strengthens the shell of Norwalk virus nanoparticles

This chapter is adapted from the publications: NanoLetters 11 4865-4869, (2011), 364 Micron 43 1343-1350, (2012). 365

Abstract

We investigated the influence of the protruding domain of Norwalk virus-like particles (NVLP) on its overall structural and mechanical stability. Deletion of the protruding domain yields smooth mutant particles and our AFM nanoindentation measurements show a surprisingly different indentation responses for these particles. Notably, the brittle behaviour of the NVLP as compared to the plastic behaviour of the mutant reveals that the protruding domain drastically changes the capsid’s material properties. We conclude that the protruding domain introduces prestress, thereby increasing the stiffness of the NVLP and effectively stabilising the viral nanoparticles. Our results exemplify the variety of methods that nature has explored to improve the mechanical properties of viral capsids, which in turn provides new insights for developing rationally designed, self-assembled nano-devices. Introduction

5.1 Introduction

Viruses are increasingly being used as functional nanoparticles in medicine and nanotechnology. Applications include the use of viruses as nanoreactors for material synthesis, 352 as building blocks in self-assembly and patterning approaches, 266,366 as nanocontainers for drug delivery, and as platforms for imaging and tumour targeting. 367 However, acquiring a thorough knowledge of their physical properties is still in development, and in order to fully exploit the vast potential that viral nanoparticles offer, it is essential to elucidate exactly these properties. 327 Recently developed Atomic Force Microscopy (AFM) nanoindentation experiments are starting to shed light on the material properties of single viral particles, including their Young’s modulus, breaking force and resistance to material fatigue. 44,169 These characteristics differ significantly between viruses. For instance, their elastic properties range from being as flexible as rubber to as hard as plexiglass. Until now, only viruses with a relatively smooth outer surface have been probed, which generally follow the predictions of thin shell elastic continuum theory. 44 However, it is unclear how a heterogeneous surface morphology of the viral particles affects their overall material properties. To study these effects, we have looked at the mechanical properties of the human pathogenic Norwalk virus, which exhibits uneven surface features with distinct hollows and protrusions. Knowledge on its mechanical properties will help us better understand general design principles of icosahedral viruses and will shed light on the binding mechanism of the virus to its host cells. The former will support the rational design of self-assembled nanoparticles whereas the latter will help in developing drugs targeted at disabling or blocking the receptor binding step of the infection cycle. Norwalk virus is a non-enveloped, icosahedral virus. Its 7654-nucleotide single- stranded RNA genome encodes for various proteins including the major capsid protein VP1 and the minor protein VP2. 244,368 Expressing the capsid proteins in insect cells with a recombinant baculovirus yields self-assembled Norwalk Virus-Like Particle (NVLP) with similar morphology and antigenicity as the native Norwalk virus. 223,245,369 VP1 folds into two distinct domains of the capsid: the S (shell) domain forms the smooth icosahedral shell while the P (protruding) domain forms protrusions that radially-project outward from the shell, creating cup-like depressions at the 3- and 5-fold symmetry axes, as illustrated in figure 5.1(A, B). The NVLPs are composed of 180 copies of VP1 with a diameter of 38 nm and exhibiting T=3 icosahedral symmetry. 121 VP2 is present in very small amounts; probably only a few molecules per capsid. Although it is suspected to be internal in the capsid layer, its precise location is still uncertain and it is not seen as part of the capsid in the X-ray crystallographic structure. 121,246

 5 Prestress in Norwalk virus nanoparticles

Figure 5.1 Electron microscopy reconstructions of the outer surface and cross section of A,B) NVLP, and C,D) CT303 particles. Hollows at the 3- and 5-fold symmetry axes are depicted in A). Modiﬁed from Bertolotti-Ciarlet et al., 245 reproduced with permission (copyright © 2002, American Society for Microbiology).

The structural requirements for assembly of the Norwalk virus capsid were investigated by mutational analyses on VP1. Complete deletion of the protruding domain leads to the assembly of mutated particles (CT303) which have smooth T=3 icosahedral shells with a diameter of ∼27 nm. 245 The structure of the CT303 particles resembles the NVLP without protrusion indicating that the shell domain alone contains the structural material required for assembly of the icosahedral capsid, as illustrated in ﬁgure 5.1. However, the icosahedral shell in the CT303 particles is slightly smaller compared to the shell domain of the NVLPs. The protruding domain, which contributes to the overall diameter of the particles, is already known to mediate virus to host cell attachment. 370 However, the eﬀects of the P domain on particle stability have remained unclear. We investigated the role of the P domain on the structural stability of the capsid by comparing the mechanical properties of the particles formed from the wild type protein and the CT303 truncation mutants. This was done by nanoindentation experiments using AFM. 44 Our results suggest that the protruding domain forces the Norwalk virus shell in a state of prestress. Prestress is a feature in biological systems which is known to govern their mechanical structure including stability and integrity.

 Materials and methods

For instance, cells and in vitro actin networks that are under prestress exhibit an increased stiﬀness. 371,372 It was recently observed that elongated bacteriophages are under an anisotropic prestress. 373 The modelling of sphere-like viruses also shows that prestress is an important component in the structure of self-assembled icosahedral protein shells. 353,356 The results of our study indicates that the icosahedral NVLP is under an isotropic prestress thereby increasing the radius of the shell domain and resulting in an overall strengthening of the viral particle.

5.2 Materials and methods

5.2.1 Protein preparation

The full-length 223 and mutant 245 versions of the Norwalk virus capsid proteins were expressed in Spodoptera frugiperda (Sf9) cells using a baculovirus expression system, as described previously. 223,245 The Sf9 cells were harvested 5-7 days after the baculovirus infection and the particles were puriﬁed using centrifugation in a CsCl gradient. The assembled NVLPs were then stored in water while the CT303 particles were stored in a 200 mM sodium phosphate buﬀer (pH 7) with 250 mM NaCl. Both stock solutions were kept at 4 ◦C. The Hepatitis B virus (HBV) particles were obtained as described by Zlotnick et al. 374 The integrity of NVLP and CT303 particles were then characterized by native mass spectrometry (MS), blue native polyacrylamide gel electrophoresis (BN-PAGE) and AFM imaging. 18,53,375

5.2.2 Mass spectrometry and native gel electrophoresis

High-resolution and tandem mass spectra were recorded on a modified Q-ToF 1 instrument (Waters, Manchester, UK) in positive-ion mode. 224 Experiments on NVLP (0.2 µM capsid concentration) and CT303 (0.1 µM capsid concentration) particles were performed in 250 mM ammonium acetate buffer (pH 7). To enhance the transmission of the large Virus Like Particle (VLP) ions, Xenon was used in the collision cell at a pressure of 2×10−2 mbar. 376 The voltages and pressures were also optimized for large non-covalent protein complexes. Briefly, the capillary and cone voltages were held at 1.1 to 1.4 kV and 155 to 175 V, respectively. The voltage before the collision cell was varied from 10 to 400 V, but generally left at 50 V for the accumulation of native ESI mass spectra. Ions were introduced into the source under an elevated pressure of 10 mbar. Experiments on HBV, at 0.04 µM capsid concentration in 200 mM ammonium acetate buffer (pH 6.8) were performed as described in Uetrecht et al. 184 In addition to MS analysis, the VLPs were analysed by BN-PAGE to confirm their purity and integrity. 377 The VLPs were diluted into sample loading buffer, 50 mM

 5 Prestress in Norwalk virus nanoparticles bis-Tris (pH 7) and 5% (w/v) Serva Blue G coomassie brilliant blue G250 (serva blue), and then 25 µg of each VLP was directly loaded onto the gel. The stacking and resolving gels consisted of a 3% and 4% (w/v) acrylamide, respectively. Both gels were prepared in a 150 mM bis-Tris buffer (pH 7) with 600 mM aminohexanoic acid. The cathode buffer was prepared just prior to analysis and consisted of a 50 mM bis-Tris buffer (pH 7) with 0.02% (w/v) serva blue. The anode buffer was 50 mM bis-Tris buffer (pH 7). Electrophoresis was carried out at 4 ◦C and 25 V for half an hour, then 150 V for a further 6 hours.

5.2.3 Atomic force microscopy

Prior to AFM measurements, the NVLP sample was diluted to 14 µg/ml and the CT303 sample to 1 µg/ml, both in buffer solution (200 mM sodium phosphate, 500 mM sodium chloride, pH 6.2). A droplet of 100 µL of the NVLP or the CT303 particles was deposited onto a hydrophobic glass cover slip (hexamethyldisilazane- treated) 169 and incubated for 20 min before the addition of 100 µL buffer solution. The particles were imaged 18,378 and nanoindented 44 with an AFM from Nanotec Electronica (Madrid, Spain), operated in jumping mode. 94 Rectangular Silicon ni- tride cantilevers (Olympus OMCL-RC800PSA) with nominal tip radii of 20 nm were used and calibrated using the Sader method yielding an average spring constant of 50.9 ± 0.2 pN/nm. 117 A series of zoomed-out images are performed on the sample to determine the particle distribution and density. An isolated particle is selected and zoomed-in. After a high-resolution image of the particle is taken, the AFM tip is positioned at the centre of the particle. The tip is locked in position, ceases scanning in x− and y−directions and starts the nanoindentation cycles by pushing on the particle in the z−direction at an average loading speed of 50 nm/s and step size of 0.4 nm. Five successive indentations are performed and the approach/retraction force-distance curves are recorded. Finally, the particle is imaged again providing visual inspection of the structural state of the particle after the indentation. Stated error, unless specified otherwise, is the Standard Error of the Mean (SEM).

5.2.4 Nanoindentation data analysis

The nanoindentation data analysis was performed using a home-written LabVIEW (National Instruments, USA) program. During nanoindentation, the parameters obtained from the AFM are the piezo displacement (in nanometers, nm) and the deﬂec- tion of the cantilever as measured by the photodiode (in volts, V). The deﬂection of the cantilever is converted to units of force (in nanoNewton, nN) by calculating the

 Materials and methods calibration factor, CF. This factor is obtained by performing a force-distance curve on a hard surface such as the glass substrate (glass curve). The slope of the linear part of the glass curve represents the deflection sensitivity of the cantilever, S (in V/nm). The calibration factor can be calculated using CF = kcant/S (in nN/V), where kcant is the spring constant of the cantilever. The force-distance curves (e.g. nanoindentation curves on viral particles) can now be expressed in force units (nN) by multiplying the cantilever deflection with the calibration factor, as shown in figure 5.2A. The measured force-distance curve is a coupling of the cantilever’s deflection and the particle’s indentation. By approximating the virus-cantilever system as two springs in series, the spring constant of the virus can be measured using the equation −1 −1 −1 keff = kcant + kvir , where keff is the slope of the measured force distance curve. The calculation for the cantilever’s deflection sensitivity is performed semiautomatically. An initial fit is performed automatically as described below, while all parameters remain manually adjustable if necessary. First a fit of the baseline of the curve is calculated by selecting a few starting points. Using the y-intercept of this fit, the entire force distance curve is translated along the y-axis such that the baseline lies at zero. Then, the starting point when the curve starts to rise is selected by evaluating the next point iteratively using either of the three options: (a) higher than the noise level, e.g. its value is twice or thrice the standard deviation of the baseline, (b) above a set force threshold, or (c) if after that point is added, the slope of the baseline changed to a certain degree from zero. If the criterion is not fulfilled, the investigated point is added to the calculation of the linear fit and standard deviation

Figure 5.2 Steps in determining the indentation curve of the virus. A) The photodiode signal is converted from volts to nanoNewton. In this example, the calculated deﬂection sensitivity (S) is 0.0248 V/nm. Using a cantilever spring constant of 0.0531 N/m, this yields a force calibration factor of 2.14 nN/V. The corresponding force is indicated in the right axis. B) The virus curve is translated such that its reference point (0 indentation) coincides with that of the glass curve. The diﬀerence between their displacements is the actual indentation of the virus, as indicated by the double arrows.

 5 Prestress in Norwalk virus nanoparticles of the baseline. As the obtained fits can be affected by noise, the programme offers the additional option to set parameters manually for individual curves. When the starting point is determined, a linear fit is performed on the rest of the glass curve as shown by the dashed black curve in figure 5.2A. The slope of the fit yields the cantilever’s deflection sensitivity. The calculation of keff is also performed semiautomatically. The starting point is determined just as in the glass curve as described above. However, in contrast to the glass curve, discontinuities or other nonlinearities appear in this curve such as due to breakage or plastic deformation. There are two semi-automated methods to obtain these parameters from the curve: (a) evaluating the force drop between two points and compare to a set threshold, or (b) from the starting point a slope is calculated from the next 5 or 10 points, then another point is added iteratively until a certain change in the slope is measured. Whereas method (a) works well for failure (i.e. generally a large sudden drop in the force), it works less efficiently for plastic deformation as the change in slope can be very gradual. Method (b) works for both types of irreversible deformation. The critical force is determined as the force where the linearity ends. In all the automated curve fitting algorithms, the results are readily observed and the fitting parameters can easily be changed in real time. As with the glass curve, the start and end points of the linear fit and the breaking/critical force can be manually adjusted to obtain an optimal fit that excludes the noisy features in the force-distance curves from the analysis, if needed. The contribution from the cantilever’s deflection can be decoupled from the actual indentation of the virus by subtracting the displacement of the virus-cantilever curve from the glass curve. This way, the indentation curve of the virus particle alone can be visualized. For this, one needs to shift one of the curves that are shown in figure 5.2A. This is done by first aligning the coupled and glass curves at the point the tip has reached the virus or glass surfaces respectively, i.e. just before the curve starts to rise linearly. This point will be denoted as the reference point (0 indentation). This can be done automatically for the glass curve by fitting a line to the linearly rising part of the curve and another for the baseline. The reference point is their intersection point. Similarly, this can also be done to the virus-cantilever curve. Then, either one of the curves is translated along the x-axis so that their reference points align. For each force value, the difference in the displacements of the virus-cantilever and glass curves is evaluated as illustrated by the double arrows in figure 5.2B. The resulting curve is the nanoindentation force-distance curve of the virus alone.

 Results and discussion

5.3 Results and discussion

5.3.1 Native mass spectra of the particles

Figure 5.3 shows the native mass spectra of HBV, 184 CT303 and NVLP. 122 Because the charge state of HBV is well characterized, we used its spectra to support the mass estimations for NVLP and CT303. The mass of VP1 (MW=56077 Da) and CT303 monomeric proteins (MW=24286 Da) were established using this mass spectrometry data. The molecular mass of the intact NVLP and CT303 particles were estimated to be 10.1 and 4.4 MDa, respectively; assuming that they consist of 180 protein subunits. The analysis of HBV, CT303 and NVLP by native gel electrophoresis revealed that the intact NVLP particles migrated less far than the intact CT303 particles. The CT303 particles moved slightly further than expected when compared to the similarly sized HBV T=4 capsids (4 MDa), 184 which is probably due to its very smooth outer surface. Furthermore, it was shown that the CT303 particles were only stable under a very narrow pH range (5-7), whereas in the NVLP sample there were still particles present at higher pH (8-9). 122 VP2 was neither detected by mass spectrometry nor in SDS-PAGE gels. 122 Therefore it remains unclear what its stoichiometry and function in capsid assembly is.

5.3.2 Size of the Norwalk virus-like particles

The immobilised particles were imaged using AFM in buﬀer solution. 18 Figure 5.4 shows AFM images of NVLP and CT303 particles and their height proﬁles and size distributions. The height distribution shows that the NVLPs have an average diameter of 38.3 ± 0.1 nm, almost the same size of the particle in the atomic model. 121 The size distribution of the CT303 particles also shows a uniform distribution with an average diameter of 27.9 ± 0.2 nm. This value is also comparable to the size determined from cryoelectron microscopic reconstructions. 245

5.3.3 Elastic properties of the particles

After imaging, nanoindentation experiments were performed. 44 Five successive indentation and retraction curves were recorded for each particle. Figure 5.5 shows the typical force-distance curves for NVLP and CT303 particles. The spring constant is determined from the linear part of the slope during the first indentation. Their k distributions are depicted in figure 5.6, showing that NVLPs have a much higher spring constant than the CT303 particles: 0.30 ± 0.01 N/m vs. 0.111 ± 0.004 N/m. The distribution of spring constants for the NVLPs is broader, likely reflecting the heterogeneity of its surface as compared to the smooth CT303 particles.

 5 Prestress in Norwalk virus nanoparticles

Figure 5.3 Representative native nano-electrospray mass spectra for A) HBV (0.04 µM capsid concentration) in 200 mM ammonium acetate buffer (pH 6.8), serving as comparison for the other spectra, B) CT303 (0.1 µM capsid concentration) in 250 mM ammonium acetate buffer (pH 7), and C) NVLPs (0.2 µM capsid concentration) in 250 mM ammonium acetate buffer (pH 7). D) Native gel: For each capsid, 25 µg was loaded onto a 4% (w/v) gel where they were analysed by BN-PAGE to further confirm their integrity. The bands at the bottom of the gel represent monomers/dimers.

Figure 5.4 Sample AFM images of A) NVLP and B) CT303 particles. C) Height proﬁles of the particles, which are used to estimate their sizes. These particular images record maximum heights of 39 nm and 29 nm for the NVLP and CT303 particles, respectively. D) Particle size distributions yield average values of 38.3 ± 0.1 nm and 27.9 ± 0.2 nm for the NVLP (n=84) and CT303 (n=92) particles, respectively.

 Results and discussion

Linear elasticity theory describes the 3D Young’s modulus E for a thin shell by E = αRk/t2, i.e. proportional to its spring constant k and radius R and inversely proportional to the square of its thickness t. 49 For the proportionality factor α, a value of 1 can be assumed. 169,183,358 Whereas the CT303 particles cannot be called ideally thin shelled particles (t << R), it has previously been shown that this equation still holds for viral particles with thick shells. 51,183,184 Thus we can apply this theory to get an estimate of E for the homogeneous CT303 particles. These particles are modelled as shells with thickness of 2.7 nm and effective radius of 12.2 nm, (taken from the middle between the inner and outer radii). Inserting these numbers into the thin shell equation yields an estimated 3D Young’s modulus of ∼0.2 GPa, which is comparable to HBV and CCMV 44 - two other viruses that also self-assemble around their genome as Norwalk virus does. Both sets of indentation curves in figure 5.5 show an almost vertical rise of the force on the right end of the plots, indicating that the AFM tip has reached a point where the particle cannot be squeezed any further. The CT303 particle has a diameter of about 28 nm thus, the vertical rise of the force at around 24 nm indicates an incompressible protein layer of ∼4 nm. Since the wall thickness of the particle is ∼2.7 nm (refer to figure 5.1 C,D), this indicates that the incompressible layer refers to the two opposite surfaces of the particle being squeezed together and partially deformed. On the contrary, for the NVLP the measured incompressible layer of about 5 nm leads to the conclusion that it is most likely not a double but only a single surface layer of the NVLP particle because it has a wall thickness of ∼7.5 nm (refer to figure 5.1 A,B).

Figure 5.5 Typical force-indentation curves of A) NVLP and B) CT303 particles. Initially, the force sensed by the AFM tip is zero until it touches the particle’s top surface (indicated by 0 nm in the indentation axis). As the tip pushes further onto the particle, the force gradually rises in a linear fashion from which the particle’s spring constant can be derived. The structural strengths of the particles were tested by indenting further beyond the linear regime leading to nonlinear deformation including plasticity eﬀects, buckling or breakage.

 5 Prestress in Norwalk virus nanoparticles

This indicates that the top surface has been torn or broken apart during the first indentation. Cuellar et al. 379 studied the material properties of NVLPs as a function of pH, showing an increase in compliance at alkaline conditions. Whereas this study is valuable to compare the influence of pH on the relative mechanical stability of NVLPs, it seems difficult to compare the absolute values with our results. In their study the particles were already heavily deformed during imaging (up to 20% decrease in height), 379 which could lead to premature damage even before the start of the nanoindentation experiments. This could explain the higher spring constant we have found at pH 6.

5.3.4 Buckling and breakage versus plasticity

Figure 5.5A shows that NVLPs typically buckle and break during the first indentation as demonstrated by a sharp drop in the measured force after reaching a critical point. The succeeding indentation curves (second-fifth) show no deformation for the first ∼18 nm, indicating that structural failure already occurred during the first indentation. This is a common feature also observed for other icosahedral capsids like CCMV, 183 Phage λ 186,380 and HSV-1. 138 Breakage of NVLPs occurs around a relative deformation of ∼22% of the radius, which is comparable to other studied capsids. 44 The critical force at which nonlinear deformation occurs during nanoindentation is at 1.1 ± 0.1 nN for the NVLPs, whereas for the CT303 particle, it is at 0.53 ± 0.03 nN. The indentation at the critical force is similar for both particles at 4.1 ± 0.4 nm and 4.9 ± 0.3 nm, respectively. In contrast to the NVLP, the CT303 particle does not exhibit buckling behaviour but instead shows a nonlinear continuous deformation. The particle appears to bounce back to a sphere-like shape after being fully compressed. This is illustrated in

Figure 5.6 The k distributions of NVLP and CT303 particles. The spring constant is determined from the slope of the force-distance curve before buckling/breakage or plastic deformation.

 Results and discussion

figure 5.5B where the force of the second approach curve goes up at the same position as the first. However, the slope of the second curve is significantly lower, indicating that although the particle did recover to its full height, it suffered material fatigue and affected its spring constant. The successive force-indentation curves show that the particle’s height progressively decreases after several indentations, as the curves start to go up at later positions. Furthermore, the change in the slope of the linear part of the third to fifth approach curves also shows that the decrease in spring constant continued. Comparing the indentation behavior of all CT303 particles indicates that they directly show a decrease in spring constant after the first indentation, while at the same time only ∼75% of the particles show a decrease in height. The other 25% show a decrease in height only after the second indentation. The decrease in diameter and elasticity after several indentations implies that although the particles do not clearly break, they undergo plastic deformation, as previously observed for HBV. 51,182,184 However, HBV deforms plastically above an indentation of about 60% of its radius, whereas the CT303 particles already start deforming nonlinearly above 36%. The nonlinear deformation of the CT303 particle occurs above an indentation force of 0.5 nN, while for the HBV T=3 and T=4 particles this is at 1.0 nN. 184 Moreover, the CT303 particles record a 30% decrease in height after 5 indentations whereas HBV capsids decrease only 10% in height under similar conditions. 182 These observations indicate that the CT303 particles are quite prone to plastic deformation compared to the HBV capsids.

5.3.5 Comparison to linear elasticity theory

To further investigate the observed plastic deformation and structural failure of the Norwalk viral shells we use an analysis involving the Föppl-von-Karman number (FvK) γ. 50,51 This dimensionless number emerges from linear continuum elasticity YR2 theory and is defined by γ = κ , i.e. the ratio of the two-dimensional stretching elasticity Y and bending stiffness κ of homogeneous, isotropic shells of radius R. Even though the underlying theory assumes completely reversible behaviour, it has been shown that for a wide range of FvK values, there is remarkable similarity between simulated indentation curves using finite element methods and the results from nanoindentation experiments that show plastic deformation and failure (nonre- versible processes). The simulations and experiments also show that this nonlinear deformation depends on the icosahedral orientation along which the particle is deformed. 51,185 However, the global characteristics of the curves are still similar for all indentations along the 2-, 3- and 5-fold axes. For small γ (<100), the simulated indentation curves 50,51 show linear deformation. For intermediate γ (100–1000), a nonlinear but continuous indentation behaviour (plastic deformation) is observed.

 5 Prestress in Norwalk virus nanoparticles

Figure 5.7 Comparison between simulations based on linear elasticity theory and nanoindentation experiments. a) Simulated force-distance curves using finite element methods for different γ. Mod- ified from Klug et al., 50 reproduced with permission (copyright © 2006, The American Physical Society). Set of first indentation curves for NVLP (b) and CT303 (c) particles.

Finally, for large γ (>1000), buckling events (or breakage) are predicted as shown in figure 5.7A. A comparison of the indentation curves of the NVLP and CT303 particles to simulated curves as shown in figures 5.7B and C indicates that the CT303 particle exhibits plastic deformation with γ ≈ 200 while NVLP exhibits buckling/breakage with γ>1000. Using an approximation of the FvK number γ = 12(1 ν2)(R/t)2, with Poisson’s ratio ν =0.33 51 and assuming that both particles are made up of material with the same mechanical properties implies that the NVLP should be a particle with a thinner shell than the mutant particle. This is contrary to the actual structure of these particles as illustrated in figure 5.1, i.e. the NVLP has a thicker shell than the CT303. Hence, the addition of the protruding domain affects the structural properties of the whole particle in an unexpected way, behaving as if it has a thinner shell. The measured spring constant of the NVLP is roughly 3× larger than the CT303 mutant. Linear response theory predicts that this spring constant is proportional to (κY )1/2/R. 50 Taking into account the difference in shell thickness (7.5/2.7∼3), this shows that the product of the bending modulus κ and the 2D Young’s modulus Y should increase with a factor of roughly 10 when the protruding domain is present. Assuming that only Y has changed by the presence of the protruding domain, this implies that the NVLP has an FvK of roughly γ∼2000, which coincides with what is observed by comparing its indentation curves to the simulations. However, this raises the question why Y changed this much. In order to dissect the origin of this unexpected behaviour we examined the molecular structure of the Norwalk virus capsids.

 Results and discussion

5.3.6 Prestress increased rigidity and strength of NVLP

X-ray crystallography of NVLP has shown that the dimeric contacts between protruding domains form distinct arch-like structures that surround the hollows at the 5- and quasi 6-fold axis of the T=3 capsid. 121 The C-terminal residues of the P domain are coupled via hydrogen bond interactions to the S domain residues. These bonds, observed in the crystal structure of the NVLP, seem to control the size of the viral particle as removing these bonds induces a change in size. 245 For instance, it was shown that the shell of the CT303 particles has a smaller radius than the shell domain of the NVLP particles. This finding shows that the equilibrium shape of the shell without the protruding domain is different from its shape when the protruding domain is present. Apparently the bridge-like arches lock the whole shell in a specific shape, which is not the ground state of the shell domain. The shell domain in the NVLP is being pulled in an outward direction by the protruding domain as illustrated in figure 5.8. Conversely, the shell domain pulls the protruding domain in an inward direction, as the shell domain would prefer its smaller ground state. This tightens the bonds between the protruding domains of the capsid proteins, like wedges being pressed together. It yields a whole shell under continuous prestress and the ensuing balance of forces seems to result in the observed drastically increased rigidity of the NVLP as compared to the CT303 mutant. This prestress puts the whole particle under tension making it more difficult to pull apart and therefore effectively increases its 2D Young’s modulus Y .

Figure 5.8 Balance of forces: Mechanical model of the CT303 mutant and NVLP. The red arrows in the NVLP model indicate the prestress acting on the shell domains (S) due to the presence of the protruding domains (P). The dark green arrows indicate the inward force of the shell domain acting on the protruding domain, thereby creating prestress in the latter. The grey dashed lines show the diﬀerence in size between the CT303 particles and the shell domain of the NVLP.

 5 Prestress in Norwalk virus nanoparticles

5.4 Conclusion

In this study, we have shown that the heterogeneous, rough protrusions in Norwalk Virus capsids have various functions. They are essential for successful interfacing of the virus to their host cells as they contain the appropriate binding sites, making them indispensable for infection. 370 We further demonstrated that they also have a stabilising role. We conclude that the protruding domain acts through the generation of prestress in the capsid shell that effectively increases the particle’s rigidity and strength. The CT303 particles are stable at neutral pH, however unlike the NVLPs, they completely fall apart by a slight increase in pH. This pH sensitivity would have disastrous effects for the wild-type Norwalk Virus particles as the small intestine where the cells they infect reside is alkaline. The CT303 particle is characterised by tight and close packing of β-barrels and our results indicate that such a packing leads to a flexible and ductile material. Nanoindentation experiments on viruses with a comparable packing of β-sheets, for instance the southern bean mosaic virus, will allow confirmation of this hypothesis. The presented results on the mechanics of NVLP and CT303 particles seem to indicate that the protruding domain stabilises the viral particles in an unexpected way by prestress. These novel insights provide handles for improving the material properties of engineered functional nanoparticles by introducing a uniform prestress that effectively strengthens these particles.



Chapter 6

Folding-driven secretion mechanism of autotransporters studied through the unfolding pathways of Haemoglobin protease

To be published 381

Abstract

Haemoglobin protease (Hbp) is an an autotransporter protein secreted by Escherichia coli. It scavenges heme from its host by degrading haemoglobin. The secretion of Hbp across the outer membrane (OM) is an ATP and proton gradient-independent process that is believed to be driven by sequential folding of the protein at the OM surface. To test this hypothesis, we approached the problem in reverse, by unravelling its unfolding pathways using atomic force microscopy (AFM). By functionalizing both ends of Hbp with Cys handles and using Au-coated AFM tips, we unraveled its β-helical structure and explored its unfolding landscape. The results show that Hbp unfolds in a sequential multi-step and multi-route pathways along at least six unfolding barriers. These barriers correspond to the β-helical rungs of the protein, and coincide with the positions of stacked aromatic residues that are buried inside the helical loop. We found that the β-helices near the C terminal prove to be the stronger unfolding barriers, indicating that the stacking of aromatic residues and their prevalence near the C terminal contribute to its structural and mechanical stability. We deduce, albeit in reverse, that the folding of these β-helices creates stable conformations in the protein that can provide the energy to overcome the entropic cost associated with the secretion of the unfolded protein across the OM. These results support the folding-driven secretion model of Hbp, whereby the interaction between Introduction the stacked aromatic residues drives the processive folding of the β-helical ladder initiating the translocation process and eventually pulls out the rest of the protein across the OM.

6.1 Introduction

Autotransporters are a class of extra-cellular proteins that Gram-negative bacteria use to interact with their environment or invade their host. They are present in almost all the Gram-negatives studied to date and constitute the simplest bacterial secretion system to transport a protein from the cytoplasm to the cell surface 382–386 . Autotransporters (ATs) are synthesized in the cytoplasm and contain three domains: the N-terminal signal peptide, the passenger domain that contains the functional protein and the C-terminal β-barrel-forming translocator domain as illustrated in figure 6.1. During secretion the protein has to pass through the inner membrane (IM) and outer membrane (OM) which are separated by the periplasmic space. ATs are secreted across the IM through the Sec translocation machinery and uses energy from ATP hydrolysis to drive the transport. This process is faciliated by the N-terminal signal peptide (sp) which targets the protein to the Sec sytem but is cleaved off after the protein is translocated. At the periplasm, the C-terminal translocation domain folds into a β-barrel and inserts inside the OM. 387,388 The passenger domain is then transported across the OM and proteolytically cleaved-off as shown in figure 6.2. Some AT passenger proteins remain attached to the cell surface such as pertactin from Bordetella pertussis or released from the surface such as haemoglobin protease (Hbp) from Escherichia coli. 389,390 The transport of the passenger protein across the OM is a proton gradient and ATP-independent process. 391,392 However, unlike the secretion across the IM, the actual secretion mechanism of the passenger across the OM is still under debate. 386,393 It was initially thought of that the passenger protein is partly or fully-folded in the periplasm as it is transported across the OM through the pore of the β-barrel. This was refuted by structural studies on the β-barrel translocation domain where they have shown that the pore of the barrel is narrower than the diameter of the folded protein. 387,394,395 A study using Cys-loop stalling technique, 396,397 where pairs of cysteine residues are introduced in the passenger protein to probe the directionality of

N sp β-barrel C

Figure 6.1 Autotransporters have conserved structural motif and are composed of three domains: N- terminal signal peptide (sp) domain, the passenger protein domain and the β-barrel translocation domain.

 6 Folding-driven secretion mechanism of autotransporters

Figure 6.2 Secretion model of the autotransporter system: i) ATs are synthesized in the cytoplasm, ii) transported across the IM through the Sec translocation machinery and sp is cleaved-oﬀ, iii) AT reached the periplasm, iv) the β-barrel is folded and inserted into the OM, v) the passenger’s autochaperone domain in the C-terminal is exported across the OM, possibly initiated by the Bam translocation complex, vi) the stack of β-helices are folded building the beta-helical backbone and thereby pulling, folding and transporting the rest of the protein out of the OM , vii) the passenger protein is cleaved oﬀ from the β-barrel domain. The dotted square indicate the still debated folding-driven transport of the passenger protein across the OM.

 Introduction translocation across the OM, has shown that the conserved and stable autochaperone domain in the C-terminal appears first in the cell surface and is critical in the initiation of the OM translocation. In addition, mutation of the hydrophobic aromatic residues in the C-terminal region with hydrophobic but non-aromatic residues 398 has shown that it blocks secretion of the protein. It is now widely suggested that a vectorial folding-driven transport mechanism is the secretion pathway of ATs, where the sequential folding of the β-helices starting from the C-terminal pulls the rest of the passenger protein out of the OM. 389,393 This process seems to be assisted by the Bam translocation complex which initiates the processive folding of the autochaperone domain and eventually drives the sequential translocation of the passenger protein across the OM. 393,399,400 AT passenger proteins have diverse functions, structures and sizes but they have a conserved architecture and topology. These proteins are made of a long stack of β- helices starting from the autochaperone domain in the C-terminal and the functional domains are appended thereafter. Most of these species belong to the SPATE family (Serine Protease Autotransporters of Enterobactericeae) which includes pertactin (Prn), 401 Plasmid-encoded toxin (Pet), 402 Immunoglobulin A1 protease (IgAP) 403 and Haemoglobin protease (Hbp). 404 The prevalence of the β-helical backbone in autotransporters indicates that they play a crucial role in their biogenesis and transport. 390,396,405,406 The conserved auotchaperone and regular stacking of β-helices at the C terminal strongly suggest that the C→N folding of the β-helical backbone provides a force to pull the rest of the protein out of the OM. We aim to investigate this mechanism using Hbp as our model system. Haemoglobin protease (Hbp) is an autotransporter which is produced by E. coli during intra-abdominal infections and promotes occurrence of peritonitis. 407 It is one of the first autotransporter and the largest β-helical structure solved to date 404 and recently the structure of its β-barrel translocation domain has just been made available. 408 The passenger domain, which we now refer to from now on as Hbp, is composed of 24 rungs of β-helices and has a cartesian length between its C and N terminals of 13 nm. The autochaperone domain in the C–terminal contains a partly disordered cap region and the first helical rung of the β-helical backbone. On top of this domain starts a series of fourteen β-helices whose core is filled with hydrophobic and aromatic residues. The first five β-helices have aromatic residues (Trp, Phe) which are stacked directly on top of each other. At two-thirds of the length, a chitinase b-like domain protrudes from the side, then another series of not so ordered beta helices is added and finally towards the N-terminal, the globular protease domain is appended. The structure of Hbp is illustrated in figure 6.3 (wild type).

 6 Folding-driven secretion mechanism of autotransporters

13 nm

wild type N256 A558 L805 Figure 6.3 Hbp wild type and truncation mutants. The diﬀerent domains of the protein are: the C-terminal autochaperone domain (cyan), β-helical backbone (green), chitinase b-like domain (blue), α helices (yellow) and the protease domain (red). The positions of the cysteine handles are indicated in magenta.

In this study, we investigated the unfolding pathways of Hbp using atomic force microscopy (AFM) to determine, albeit in reverse, if the folding of the β-helical backbone is capable of pulling the rest of the protein out of the OM. Hbp constructs were designed to speciﬁcally probe the mechanics of unfolding of the β-helical stacks and to examine which interactions between residues give rise to unfolding barriers. The unfolding pathways provide insights in the stability of the diﬀerent parts of the folded regions and the magnitude of energy stored in the folded molecule. In addition, we performed rough calculations based on the contacts between residues and hydrogen bonds to predict which structure in the truncates are vulnerable or strong against mechanical stretching and unfolding. Understanding the mechanics of the autotransporter secretion pathway across the OM solves an open fundamental research question but is also useful in the area of biotechnology where the autotransporter secretion pathway is becoming a promising tool for high-level, high-capacity and high-throughput platform for expression and surface display of recombinant proteins. 409–413

 Materials and Methods

6.2 Materials and Methods

In this study, we ﬁrst focused on the smallest construct to obtain simpler unfolding pathways and to better understand what initiates the vectorial C→N folding-driven transport. We describe the cloning and production of this truncation design for Hbp. This is followed by the description of the mechanical unfolding experiments using atomic force microscopy (AFM) and the description of the computational method used to estimate the energy needed in unfolding the protein structure from the number of contacts and hydrogen bonds broken.

6.2.1 Cloning and production of Hbp truncation mutants

We used the Hbp genes that encode the wild type and three truncated Hbp passenger domains starting at positions N256, A558 and L805 with respect to the start of the wild type passenger. The numbering is based upon the sequence of the Hbp passenger structure (PDB code: 1wxr). 404 The cloning of the truncated constructs and the secretion of the resulting passengers has been described by Jong et al. 414 In all constructs the truncated passengers are preceded by the short linker GSSCGSGSG. The linker contains a cysteine residue that serves as a handle at the N terminal for the AFM experiments (indicated in magenta in figure 6.3). To include a similar Cys-residue handle in the C-terminal part of the passenger, a cysteine was engineered at position 967 of the passenger, substituting the serine residue at that position. The cloning involved a three-primer PCR strategy using Phusion DNA polymerase (Finnzymes). In a first PCR reaction, primers Hbp- mut S967C (5’-CAAGCGTGTCCTTGTTACAAGGTTTT-3’) and pEH3-Rev (5’- GTGAATTCGGATCCAGAGATGTGT-3’) were combined with pEH3-Hbp plasmid as DNA template. 414 The resulting amplicon was purified by gel extraction (Qiagen) and used in a second PCR in combination with primer Hbp-KpnI-F (CGGGTACCG- CAATATCTGGA) and vector pEH3-Hbp as a template. The resulting amplicon was cloned into the vectors that encoded the truncated Hbp using restriction enzymes KpnI and EcoRI. The resulting plasmids were checked by sequencing for the presence of the mutation that results in the S967C substitution. The introduction of the linkers and the cysteine residues did not affect secretion and processing of the Hbp passenger. The plasmids encoding the wild type and truncated Hbp proteins with their cysteine handles were introduced in Escherichia coli strain Top10F (Invitrogen) for protein production and secretion. Cultures were grown in 100 mL Lysogeny Broth (LB) and expression of the Hbp constructs was induced by adding 0.5 mM IPTG. Two hours after induction, the cultures were placed on ice and all subsequent steps were

 6 Folding-driven secretion mechanism of autotransporters

A B

Figure 6.4 A) Coomassie-stained SDS-PAGE gels of cells (pellet) and culture supernatants (sup) of E. coli cultures expressing the constructs for wild type Hbp and the truncated mutants N256, A558 and L805. Cultures were centrifuged to obtain the cells and the supernatants. Prior to loading onto gel, the supernatants were concentrated 10× compared to pellet samples by precipitation with trichloroacteic acid. Expression and secretion of Hbp and its mutant derivatives results in secretion of the passenger part of the protein, with some passenger remaining in the cell pellet 414 (•). Indicated further are the positions of the β-domain (C) and the position of non-cleaved β-domain-passenger fusion for the N256 and L806 truncates (J), which resulted from ineﬃcient cleavage of the β-domain. B) Coomassie-stained SDS-PAGE of the concentrated protein samples used for AFM. Loaded were un-diluted samples of 5-10 mg/mL. The disulﬁde bridge formation between the cysteine residues in the truncated Hbp mutants cause the formation of some multimers (indicated by •), but this can be resolved by a short incubation with dithiothreitol (DTT). Wild type Hbp does not contain cysteine residues.

performed at 4◦C. First, cells were pelleted from the culture by centrifugation (1 400g for 10 min). The supernatants were collected in new tubes and centrifuged at 10 000g for 30 min and filtered through 0.2 µm filters (Millipore) to remove residual cells. Sub- sequently, the supernatant was concentrated and purified by sequentually passing it through centrifugal concentrators. First a concentrator with a 150 000 kDa cut-off (Invitrogen) was used, after which the filtrate was concentrated using concentrators (Vivaspin20, Sartorius) with cut-offs of 30 000 kDa (wild type, N256 and A558 truncates) or 10 000 kDa (L805 truncate). When the sample was concentrated to 1 mL, the medium was exchanged for 10 mM PBS (pH 7.4) by diluting the sample in PBS to 6 mL and subsequent concentration to 1 mL using Vivaspin6 concentrators and this was repeated five times. The concentrators used were centrifuged at the maximal speeds indicated by the manufacturer, in runs of 20-30 minutes. The samples were analysed by SDS-PAGE and stained with Coomassie Brilliant Blue. On SDS-PAGE the samples appeared to contain over 90% of Hbp with only trace amounts of other proteins being detectable as shown in figure 6.4.

 Materials and Methods

The truncates were designed such that the unfolding of interesting structures in the protein, e.g. the β-helical backbone, can be investigated. For instance, the shortest construct (L805) contains the ﬁrst stack of six parallel β-helices after the autochaperone domain from the C-terminal. Each helix is composed of two β strands (long and short) which form a loop with a triangular cross-section. This part contains largely the hydrophobic and aromatic residues in the protein, which are mostly buried inside the loops and stacked directly on top of each other. Recent studies have suggested that the Bam complex assisted the translocation of the autochaperone domain but the folding of the succeeding β-helical stack initiates and eventually drives the transport of the entire passenger protein across the OM. 399 Using this truncate, we will determine if there is energy gained in the folding of these helices by investigating the presence of unfolding barriers. In addition we also investigate the presence of stacked aromatic residues in connection to the position of the barriers. The intermediate truncate (A558) includes the next stack of eight β-helices. The loops of these β-helices are rotated by about 60◦ relative to the loops in L805 and there is less stacking of hydrophobic and aromatic residues inside. Finally the largest truncate N256 contains the whole of the β-helical backbone except the protease domain. It comprises the chitinase b-like domain and the last stack of ten β-helices which are less ordered than the ones before them.

6.2.2 Imaging and mechanical unfolding using AFM

Single molecule imaging and manipulation using atomic force microscopy has emerged as a very powerful tool in biophysics. The combined capability of AFM to image small molecules which are below the diffraction limits of light-based microscopy, and to perform mechanical manipulation and force spectroscopy on these molecules have opened up new and interesting research areas, such as the study of DNA-protein interactions, 9,129,148,415,416 probing the structural strength and mechanical properties of viruses and cells, 18,44,417–419 elucidating the binding strengths of molecules such as antibody/receptor 418,420 and unravelling the folding/unfolding mechanics of proteins. 52,80,163,177 AFM has found a special niche in protein unfolding experiments because it can mechanically manipulate the proteins in physiological buffer conditions. 54,55 Moreover, it is relatively easy to bind small proteins to the surface or the AFM tip without the use of long handles as is often needed in other single-molecule techniques such as optical or magnetic tweezers . 10,65,68 In unfolding experiments, single- or multi-domain proteins are functionally-attached to the surface and probed by the AFM tip using specific or nonspecific binding. When the tip pulls away from the surface it stretches and elongates each domain of the protein, whereby at a certain force a molecular bond

 6 Folding-driven secretion mechanism of autotransporters breaks or a structural element unfolds resulting in a sudden increase of the contour length and a drop of the force. 80,163 The first unfolding experiment using AFM was the repeated stretching of the giant muscle protein titin, exhibiting reversible unfolding of the individual immunoglobulin domains. 175 Other experiments include unfolding the proteins EYFP, 214 GFP, 215 calmodulin, 59 fibrinogen, 421 α-helical stacks of ankyrin repeats, 422 among others. Simulations and models have sprung up to explain these unfolding experiments and revealed insights into the exact molecular dynamics that take place, 423 such as the presence of intermediate unfolding states due to rupture of hydrogen bonds in titin domains, 217,218 titin kinase 213,219 and GFP. 80,220,424 In this study, the AFM measurements were performed on a Bruker BioCatalyst sytem using the Peak Force Tapping™mode. For imaging purposes, SiN tips (Olympus OMCL-RC800PSA) with nominal spring constants of 50 pN/nm were used. The Hbp samples were diluted in 5 mM Tris + 5 mM MgCl2 buffer solution and incubated for 20 minutes onto a freshly-cleaved mica surface. The Mg2+ ions in the buffer facilitated the attachment of the proteins on mica. The samples were diluted up to 20× (protein concentration of 1∼2 mg/mL) until single proteins were seen isolated from each other. The Hbp samples were imaged at resolutions of 4 to 2 nm/pixel. Using these images, the size of the particles were determined.

x (nm) y (nm)

Figure 6.5 AFM setup. The Hbp constructs are immobilized on the surface through Au-S bond between one of its cysteine handle and the Au-coated surface. An Au-coated AFM tip ﬁshes out the other end of the protein via the other cysteine handle.

 Materials and Methods

For the unfolding experiments, Au-coated SiN tips (Olympus BL-RC-150VB-C1) with a nominal spring constant of 30 pN/nm were used. These cantilevers were calibrated before any experiment were performed using the thermal tuning method. 115 The Hbp samples were diluted in 50 µL PBS buffer at pH 7 to a concentration of 1∼2 mg/mL and incubated onto an Au-coated glass cover slip. The proteins were immobilized on the surface through the covalent Au-S bonding between the gold surface and cysteine handles of the protein. 68,425 After 20 minutes, the sample was rinsed five times with buffer solution to remove unbound proteins and finally added with 100 µL of buffer. The tip performs 10 to 20 approach/retract cycles at each xy position, spaced at least 25 nm apart so as not to pull the same protein molecule, as illustrated in figure 6.5. This yielded an unfolding efficiency of about 5%, but barely 1% resulted to full-length unfolding of the protein. For example, we performed about 40,000 approach/retract cycles for the L805 construct and only 250 curves show complete full-length unfolding. In this study, we only analysed the curves that resulted to full-length unfolding of the protein.

6.2.3 Image and force-distance curve analyses

Programs written in LabVIEW™were used for the image and unfolding data analyses to measure the size of the Hbp constructs and analyze the force-distance curves of the unfolding events. The image analysis performs flattening (corrects for z-tilt), despiking (removes glitches), thresholding (distinguishes protein particles from the background), particle labeling and height measurement. The size of the proteins were determined from the maximum height (highest pixel value) of each particle. The unfolding data analysis involves curve selection (of unfolding event), cantilever bending correction (subtracts the contribution from the bending of the cantilever), curve alignment and Worm-Like Chain model fitting of each unfolding fragment. 426–428 To facilitate the superposition of the unfolding curves and correct for unspecific events at the start of the unfolding experiments, the curves were translated in the x- axis such that the last unfolding event is aligned with the fully-unfolded length of the protein. The linear unfolded length is computed by multiplying the number of residues by 0.35 nm, the approximate length of each amino acid. For instance the L805, A558 and N256 constructs which are composed of 162, 409 and 711 residues (between their cysteine handles) have approximate fully-unfolded lengths of 57, 143 and 250 nm, respectively. The translation process also determines the selection criteria for the unfolding curves to be used in the analysis. The positive translation is bounded by the measured folded size of the proteins, e.g. ∼4 nm for L805 while the negative translation is bounded by twice their size due to additional unspecific binding effects

 6 Folding-driven secretion mechanism of autotransporters such as adhesion. The curves which record the correct protein length and satisfy the above criteria are further analyzed. Figure 6.6 shows a sample of the force-distance curves measured during the unfolding experiments. These curves record the force applied on the AFM tip as it approaches and retracts from the surface. The retraction curve illustrates an event in which an Hbp molecule attaches to the tip during approach and fully unfolds as the tip is pulled away from the surface. The unfolding of the protein alone is extracted from this curve by deleting the baseline and linear parts and correcting for

A B vi v

iii ii i ⇨ ⇨

D C

⇨

Figure 6.6 A) Force-distance curves recorded during an approach-retract cycle. B) The retraction curve shows the force needed to pull and unfold the protein. The illustrations show how the pulled protein molecule stretches, then encounters a barrier and finally unfolds. During these instances, the force on the cantilever is gradually increasing (i), steeply increasing (ii, v, vi) and finally dropped drastically as the barrier is unfolded (iii, iv). C) The unfolding curve of the protein alone (cantilever bending corrected) shows multiple unfolding events. These events stem from the multiple unfolding barriers in the protein resulting to a multiple number of unfolding fragments. The maximum force at which the fragment unfolds is indicated by the circle in magenta. To determine its contour length (fully unfolded length), each fragment is fitted with a Worm-Like Chain (WLC) model with a persistence length of 0.4 nm. The unfolding curve is then translated in the x-axis such that the last unfolding fragment correspond to the fully-unfolded length of the protein, e.g. 57 nm for the L805 construct. D) The sample distributions for the contour length, Lc and maximum forces of the unfolded fragments in C.

 Materials and Methods the bending of the cantilever as illustrated in ﬁgure 6.6C. This correction is done by transforming the x-axis to xunfold = xretract − F/k, where F and k are the recorded force and the spring constant of the cantilever respectively. The curve may display multiple distinct sections as the protein unfolds. These sections refer to local barriers or parts in the protein that are hard to unravel. Each fragment is then ﬁtted with the Worm-Like Chain (WLC) model

−2 kB T 1 x x 1 F = 1 − + − (6.1) Lp 4 Lc Lc 4 where kB , T , Lp and Lc are the Boltzmann’s constant, temperature, persistence and contour lengths of the protein. The WLC model has been shown to describe the elastic behavior of molecules such as DNA, RNA and proteins. 423,429 The persistence length is a measure for the elasticity of the molecule and the contour length describes its full unstretchable length, which is indicated by the position where the force increases sharply while the protein is pulled. Due to thermal fluctuations and other stochastic processes, the protein can unravel at lower forces. By fitting the unfolding curves with a known persistence length (approximately 0.4 nm for proteins), the contour length of the unfolded protein can be recovered. For multiple unfolding events, the contour length corresponds to the unfolded length of the fragment and thus the position of the unfolding barrier. The exact position of the reference point (zero position) of unfolding is sometimes ambiguous due to adhesion effects or the inaccuracies of the algorithm used to cut-off the linear part of the force-distance curve. This inaccuracy is minimized by finding a model that represents the barriers of the protein. This is done by looking for groupings in the Lc distribution. Each unfolding curve is then translated by an amount which maximizes its cross-correlation with the model curve. This procedure is illustrated in the Appendix, figure 6.20.

6.2.4 Computational estimate of unfolding parameters

The energies of unfolding specific fragments of Hbp were estimated based on the number of contacts and hydrogen bonds broken. Two residues make a contact if their Cβ atoms lie within 7 Å of each other while the pairwise hydrogen bonds in the peptide backbone are assigned by the DSSP program based on the given PDB structure. 430 For each fragment to unfold, three different groups of contacts need to be broken (either Cβ or hydrogen bonding contacts): internal contacts, external contacts and cross contacts. An internal contact is defined as the contact between two residues that lie within the fragment to unfold. An external contact is a contact between one residue that lies within the fragment and one outside the fragment. A cross contact

 6 Folding-driven secretion mechanism of autotransporters is defined as the contact between residues which are outside the fragment to unfold, i.e. a residue in the N-terminal region prior to the fragment and a residue in the C- terminal region after the fragment. These contacts are illustrated by the red broken lines in figure 6.7A. During the unfolding of a fragment, all of the above contacts need to be broken. The unfolding energy for a fragment is estimated by the number of contacts broken, the total number of hydrogen bonds broken or a combination of these two. To calculate the recovered length of a fragment we consider the new length of the structure on which we pull, minus the old length of the structure. The old length is simply obtained from the cartesian distance (shortest linear distance) between the Cα coordinates of the cysteine handles on which the pulling experiment is performed (a→d). The new length is calculated by the sequence length of the unfolded fragment (b→c) multiplied by 3.5 Å per residue as the length estimate of the unfolded chain. The parts of the protein that did not unfold but changed its configuration/orientation after unfolding need also to be taken into account, e.g. between the N-terminal cysteine handle and N-terminal fragment residue (a→b), and the C-terminal pulling residue and the C-terminal fragment cysteine handle (c→d). These lengths are taken from the 3D coordinates in the crystal structure. Note that these lengths (unfolded

A a B

before after unfolding b ii a iii i

iii b c ii c i

d d

Figure 6.7 Calculation of the estimated energies in unfolding specific fragments in the L805 construct. A) The estimated unfolding energy based on the number of contacts, hydrophobic contacts, hydrogen bonds, and a combined energy from contacts plus twice the hydrogen bonds. A contact is defined as i) internal, ii) external and iii) cross contacts as illustrated by the red broken lines. The measurement of the unfolded length is the difference between the length of the structure before (a→d) and after unfolding (a→b→c→d) as indicated by the green lines. B) The derivative or gradient plots of A (change in the number of contacts or hydrogen bonds with change in fragment length), showing peaks which correspond to breaking large chunks of contacts or hydrogen bonds in unfolding a specific fragment.

 Results and Discussion fragment length and the lengths of the two structured regions) may simply be added as the angles will be optimised by the pulling. For each possible fragment, the unfolded length and the unfolding energies were calculated as described above. All possible fragments within the protein sequence that have unfolded lengths within a 1 nm window were considered. From this list, the specific sequence which has the minimal unfolding energy, based on either the contacts or hydrogen-bonding based energies is chosen. Thus, each point in figure 6.7B corresponds to a specific sequence in the protein that has been unfolded. The search for these fragments were made everywhere throughout the protein as long as the residues are contiguous. Sequential unfolding that resulted from the calculation were tested by fixing the start of the fragments considered (either N or C terminal) and compared with those that started randomly in the protein.

6.3 Results and Discussion

First, the imaging results that determine the particles sizes of the diﬀerent Hbp constructs are discussed and compared to the solved crystal structure of Hbp. Then, the results of the unfolding experiments from the shortest construct (L805) are investigated in detail because it can reveal how autotransport across the OM starts from the C terminal in a simpler way. The unfolding curves found display very rich and complex unfolding characteristics and are compared with the results from the calculations. From this analysis, structures corresponding to unfolding barriers are revealed and subsequently, a model of the unfolding pathways and energy landscape of L805 is created. Finally, the unfolding results for the intermediate (A558) and long (N256) constructs are discussed and compared with the interpretations obtained from L805.

6.3.1 Particle size distribution of the Hbp constructs

A sample AFM image of Hbp proteins deposited on mica is illustrated in figure 6.8. Using this kind of image, the size of each protein molecule was determined from the maximum of its height profile as shown. The size distributions of the Hbp constructs are shown in figure 6.9. The AFM image shows that most of the proteins adhere to the surface in an upright position. The shortest and intermediate constructs (L805 and A558), which are mostly composed of β-helical stacks were found to be rigid with average heights of 3.6 nm and 8.1 nm, respectively. These sizes correspond closely to 1 2 404 the expected ⁄3 and ⁄3 of the wild type structure as solved by x-ray crystallography. On the other hand, the N256 construct displayed a bimodal size distribution.

 6 Folding-driven secretion mechanism of autotransporters

The first peak of N256 coincides with the size of the A558 (8 nm) and its second peak fits closely to the size of the wild type structure (13 nm) as measured from x-ray crystallography. 404 Interestingly, the size distribution of the wild type exhibits the same bimodal distribution as N256 (figure 6.9 inset). These results suggest that there is a flexible hinge in the protein at the N terminal position of the A558 construct, i.e. the β-helical stack from the C-terminal (up to A558) forms a rigid structure, and above it up to the N-terminal region which comprises the protease and chitinase domains is floppy or hinged. This could be expected as the part of the β-helical backbone that extends from where the A558 construct is truncated, is less ordered and the stacking of the beta strands do not align on top of each other. We think that the way the protein can adhere to the surface might reveal the structure’s floppiness or hinge, i.e. whether it is attached via its C- or N-terminal end. Those deposited via its N-terminal side will appear smaller (due to squeezing of the floppy region or bending of the hinged region) while those attached on its C-terminal side will show the full height. Indeed, in both cases (N256 and wild type), the two populations occur at the same probability (50:50). This orientation dependence has not affected L805 and A558, thereby supporting the inference that these structures are indeed rigid.

Figure 6.8 AFM image of Hbp proteins (N256 construct) and the height profiles of individual molecules. The dotted lines indicate where the height profiles are taken. The size of each protein is determined from the maximum of its height profile. In the image analysis, the size is determined from the maximum pixel of each particle.

 Results and Discussion

Figure 6.9 Particle size distributions. The shortest (L805) and intermediate (A558) constructs show uniform distributions while the N256 construct and the wild type show two populations (inset). This implies that L805 and A558 form rigid structures, while N256 and wild type has a ﬂexible region close to the N-terminal. Measured sizes of the diﬀerent constructs are: L805 = 3.6±0.1 nm, A558 = 8.1±0.1 nm, N256 = 8.0±0.1 nm, 12.3±0.1 nm and wild type = 9.0±0.2 nm, 13.6±0.3 nm. The dotted line at 13 nm indicates the size of the wild type protein as measured from x-ray crystallography.

6.3.2 Multi-step and multi-route unfolding pathways of Hbp

The unfolding experiments reveal that Hbp does not unfold in a single event but rather in multiple discrete steps. Figure 6.10 shows a summary of unfolding curves exhibiting two-step unfolding pathways which represent 42% of the recorded unfolding events. In a 2-step pathway, the unfolding curve shows one sharp drop in force before it gets fully unfolded and breaks off from the AFM tip. The maximum force and fragment length at which this happens for each 2-step unfolding curve is indicated by either black or blue data point and the complete unfolding by gray, respectively. The plot also shows that there is not one but multiple routes for the 2-step unfolding pathway. This result illustrates that the unfolding landscape of Hbp is a complex multi-step and multi-route and energetically quite flat landscape even for the shortest construct. This is different from what is observed for other mechanically-unfolded proteins which often exhibit one-step unfolding for each monomeric unit of the protein such as titin 175 and ubiquitin. 431 Mechanical unfolding of bigger proteins reported the presence of intermediate states such as bifurcated unfolding pathways in GFP 432 and four-intermediate unfolding states in calmodulin, 433 but these proteins do not exhibit the multi-route pathways of Hbp.

 6 Folding-driven secretion mechanism of autotransporters

Figure 6.10 Summary of unfolding curves exhibiting 2-step unfolding pathway for the L805 construct. A) Each blue/green data point indicates the maximum force and fragment length at which the protein unfolds halfway while the black data point indicates when it fully unfolds and breaks oﬀ from the AFM tip. The rest of the data points in the force-distance curves are not shown for clarity. B) Using the WLC model with Lp = 0.4 nm, the contour length of each unfolded fragment length was determined. The contour length distribution indicates the presence of peaks at Lc = 7, 16, 25, 33, 41, 47 nm. The WLC curves of these groups are indicated by red lines in A. C) The distribution of the maximum forces recorded during the unfolding of these fragments are shown on the left (in cyan) and including the full unfolding as it breaks oﬀ from the tip (in black).

Figure 6.11 Six routes for the 2-step unfolding pathway. The routes that encounter the unfolding barriers at 41 and 47 nm are more probable (about 40%) and record higher unfolding forces. This indicates that the structures encountered in unfolding these fragments strongly resist unfolding.

 Results and Discussion

Figure 6.12 N-step unfolding pathways. A) The distribution of the contour lengths of the unfolding fragments for the L805 construct, grouped according to the number of unfolding events. The presence of peaks in the overall Lc distribution indicates stable unfolding barriers in the protein at positions: 7, 16, 25, 33, 41 and 47 nm which seem to suggest the sequential unfolding of the β-helical stacks of the protein. B), C) The distribution of maximum unfolding forces grouped according to the number of events and unfolding barriers.

Figure 6.13 Sample routes for the 3-step unfolding pathway. In this case, the protein unfolds twice (ﬁrst is indicated in blue and the second is indicated in green) before it fully unfolds and breaks oﬀ from the tip (in black), producing three unfolded fragments. The routes show that about 50% of the time, the protein unfolds last and records higher forces at the 41 nm barrier.

 6 Folding-driven secretion mechanism of autotransporters

In spite of the apparent complexity, the distribution of the measured contour lengths suggests that there are at least six discernible routes for the 2-step pathway as indicated by the peaks in figure 6.10B. It shows that the L805 molecule encounters either one of the six unfolding barriers and unfolds into two fragments stochastically. We assume that stably-folded sub-structures in the protein are harder to unfold and thus represent the dominant unfolding barriers. Indeed, the two most dominant barriers at 41 nm and 47 nm also show high unfolding forces. The detailed plots of these groups are shown in figure 6.11, where the barriers with shorter fragment lengths have an average unfolding force of about 80 pN while the 41 nm and 47 nm barriers display average unfolding forces of 100 pN. Aside from the 2-step unfolding pathways, we also examined the probabilities and routes of the other multi-step unfolding pathways. Figure 6.12A shows the different N-step unfolding pathways of the L805 construct and the distributions of the contour lengths and maximum forces of the unfolded fragments. The results show that the L805 molecule unfolds in single, 2−, 3−, 4− and 5−step fashions, with probabilities of 20%, 42%, 25%, 10%, and 3% respectively. Interestingly, the individual and overall distributions indicate the presence of the same prominent barriers as seen before in the 2-step pathways. This suggests that the protein unfolds by hopping stochastically between these unfolding barriers. Moreover, the lengths of the fragments (8∼10 nm) indicate unfolding a fairly regular structure in the protein, suggesting that this could be the unfolding of each helix in the β-helical backbone. The unfolding forces show skewed distribution towards the higher forces. All of the N-step pathways have similar force distributions with average unfolding forces of about ∼100 pN as shown in figure 6.12B. Measurements were performed at different loading rates from 50 to 15000 nm/s but most of the complete unfolding curves were obtained at loading speeds of 100, 200 and 500 nm/s, which represent 65%, 15% and 20% of the data, respectively. The results show that at these speeds, the distribution of unfolding forces are very similar as shown in Appendix, figure 6.21. Figure 6.12C shows the force distribution for the six unfolding barriers. The results show that the protein encounters the unfolding barriers at the same probability (∼30%) except at 7 nm (13%). The unfolding barriers at 41 nm and 47 nm record higher unfolding forces as indicated by the higher median value. This is indeed what is exhibited by the 2-step and 3-step unfolding pathways (refer to figures 6.11 and 6.13) where the barriers located at 41 nm and 47 nm show higher unfolding forces and are the most probable routes of unfolding. Based on the probability and the distribution of unfolding forces, the results indicate that these unfolding barriers have similar activation energies. 423,434,435 At the same loading speeds (100-300 nm/s), the measured unfolding forces (median values of about 100 pN) are similar to those

 Results and Discussion measured in unfolding titin, 175 fibronection 436 and GFP. 215 The mentioned unfolded proteins have force distributions which have long tails in the low force region. In contrary, the force distribution of Hbp is skewed towards higher forces at all loading speeds and these high forces occur more often at the 41 nm and 47 nm barriers. The routes shown in figures 6.11 and 6.13 also suggest that the protein unfolds sequentially among the identified barriers as illustrated by the routes of the 3-step which can be mapped on the 2-step pathways, and vice-versa. An example is the route from the 3-step unfolding pathway at 16-41-57 nm, where the two barriers encountered were the same ones found in the 2-step pathway. We aim to understand what these barriers represent and if unfolding is a sequential process. We performed calculations based on the crystal structure using the number of contacts and hydrogen bonds to obtain which parts of the protein display structural and conformational stability and compared the outcome with our experimental results.

6.3.3 Structure assignment of unfolding barriers

The presence of the same unfolding barriers in the different N-step pathways and the indication that the n-step pathways can be mapped out from the (n+1)-step pathways seem to imply that the protein exhibits an ordered unfolding process. This sequential unfolding pathway of Hbp was further examined by assigning these unfolding barriers to the actual physical structure of the protein. To do this, the number of contacts and hydrogen bonds broken during unfolding were calculated based upon the available structure of Hbp (PDB: 1wxr). 404 The whole protein construct is searched through for continuous fragments that unfold to a specific length and ranked according to the number of amino acid contacts and hydrogen bonds broken to unfold them. The chosen sequence is the one which corresponds to the lowest energy, i.e. the minimum number of contacts and hydrogen bonds. The different sequence combinations and their total energies are illustrated in figure 6.22 in the Appendix. Figure 6.14B shows the result of this calculation where peaks correspond to unfolding barriers and compared to the experimental results for the 2-step unfolding pathway of L805 as shown in figure 6.14A. In the calculations, the hydrogen bonds are weighed twice than that of the contacts. The Lc distribution of the fragments for the 2-step unfolding pathway exhibits almost one-to-one correspondence with the calculations, as highlighted by the gray band. This means that the unfolding barriers found during the experiment coincide with fragment lengths that require breaking large chunks of contacts and hydrogen bonds to unfold them. Figure 6.14D shows which structures are unfolded that correspond to the peaks in 6.14B. Interestingly, the calculation results and ranking show sequential unfolding of the protein from the N to C terminal regions as shown in figure 6.22 (Appendix).

 6 Folding-driven secretion mechanism of autotransporters

Figure 6.14 Comparison between calculation and experiment results for the 2-step unfolding pathways of L805. A) Contour length distribution of the 1st unfolding fragments (the 2nd or rather the last is the complete unfolding of the protein at 57 nm). B) The gradient plot showing the results from the calculations (contacts + 2×hydrogen bonds). C) Positions of the aromatic residues and α helix. The aromatic residues are weighted according to: 1 ≡ rings buried inside the β-helical loops, 0.5 ≡ rings outside but close to other aromatic residues and 0.2 ≡ rings outside. The stacked aromatic residues are indicated in red. D) Illustrations of the unfolded fragments (in red) that correspond to the peaks in the calculations. The positions of these unfolding barriers coincide with that of the buried aromatic residues that are stacked on top of each other as indicated by the dashed line.

This indicates that it is easier to unfold the structures from the N terminal than the structures from the C terminal region. The illustrations in figure 6.14D shows that short fragment lengths are likely the unfolding of the β-helices and α helix in the N-terminal region because they require the less number of contacts and H-bonds to break to unfold them, while the last three β-helices close to the C terminal region are the hardest to unfold. The calculated unfolding barriers represented by peaks at 7 and 18 nm correspond to unfolding parts of the first two β-helices, followed by the unfolding of the α-helix at 28 nm, and the hardest to unfold are the three β-helices at the C terminal which corresponds to the stronger peaks at 33, 39 and 47 nm. A close examination of these regions show that these unfolding barriers coincide with the positions of the aromatic residues that are buried and stacked on top of each other inside the β-helical loop, as indicated by the red lines in figure 6.14C. Their positions in the protein is illustrated by the blue line in the rightmost structure

 Results and Discussion

Figure 6.15 Comparison between calculation and experiment for all N-step unfolding pathways of L805. A) Cumulative Lc distributions of all unfolding events. B-D) Lc distributions of the ﬁrst, intermediate and last unfolding events, excluding the complete unraveling at 57 nm. E) Calculation results, F) Positions of the aromatic residues.

illustration in figure 6.14D. The protein seems to resist unfolding in positions where an aromatic residue is buried inside the β-helix, most specifically in the C-terminal region where these aromatic residues are directly stacked on top of each other. The correspondence between calculation and experiment is a strong indication that the Hbp L805 construct unfolds sequentially from the N-terminal to C-terminal and that this sequential unfolding is due to breaking one-by-one the interaction or attractive bond between the buried and stacked aromatic residues inside the β-helical core. The calculations were also compared to the experimental results for all the N-step unfolding pathways of L805. Figure 6.15 shows the contour length distribution of the first, intermediate and last events (prior to the complete unraveling of the protein at 57 nm) for all the N-step (single to 5-step) unfolding pathways. The first events distribution shows that the protein can unfold in any of the six unfolding barriers, and the last events distribution shows that the protein unfolds last in any of the four peaks from 25 nm. The intermediate events distribution shows that those unfolding in 4 or 5 steps likely unfolds first either the 1st or 2nd β-helix from the N-terminal

 6 Folding-driven secretion mechanism of autotransporters region, then unfolds the α-helix or the 3rd β-helix and lastly get stuck either at the 4th or 5th β-helices. These results indicate that the ﬁrst two β-helices are the weakest unfolding barriers of the protein and the last two are the strongest.

6.3.4 Barriers coincide with stacked aromatic residues

The protein structure is examined to investigate the conformation of the aromatic residues found at the barrier positions. A top to bottom view of the structure reveals rings of aromatic residues buried and stacked inside the β-helical loop as shown in figure 6.16. It also shows that the stacking is more regular close to the C-terminal (bottom), with six residues on top of each other. The positions of these aromatic residues correlates well to the Lc distribution of the experiments as shown in figure 6.14. Furthermore, it shows that most of the last unfolding events get stuck at the last two barriers, corresponding to the bottom two β-helices where residues F911 and F933 are located. This was also shown as the preferred last unfolding routes in almost all of the N-step unfolding pathways as illustrated in figures 6.11 and 6.13. This result indicates that the bond between the β-helical stacks is stronger close to the C-terminal. This can be explained by the close proximity of the buried aromatic residues with each other. For instance, F911 and F933 exhibit stronger resistance to unfolding because there are two/three aromatic residue stacked on top/bottom of them. The results from the experiments and calculation suggest that these aromatic residues form a cohesive bond that stabilizes and holds the β-helices together, and resists unfolding. Since the discovery of a secreted plant virulence factor made of right-handed coil of parallel tri-β strands called Pectate Lyase-C, 437 the number of proteins that exhibit the same β-helical fold have expanded dramatically. 438 The superfamilies of proteins having this β-helical backbone structure that include phage P22 tailspike endorham- nosidase 439 and pertactin 401,440 show extensive stacking of aliphatic, aromatic and polar residues. Protein structural studies have shown that the noncovalent interactions of aromatic side chains of phenylalanine, tyrosine, histidine and tryptophan prefer the orientation that enhances the π-stacking of their aromatic rings in either parallel or T-shaped configuration on top of each other. 441,442 A study of 24 pairs of structurally-similar thermophilic and mesophilic proteins have shown that significant numbers of pairwise aromatic residues are found specifically in the rigid regions of the proteins which most probably contribute to their high thermophilicity and ther- mostability. 443 In addition, recent studies on the molecular structure of the amyloid fibril, the precursor to neurodegenerative diseases caused by misfolding and aggregation of proteins, have shown that the π-bonding between aromatic rings such as that

 Results and Discussion

AB W814

W873

W891

F911

F933

F988

W1015

Figure 6.16 A) Top-down view of L805. Stack of seven aromatic residues (in cyan and blue) buried inside the β-helical backbone. B) These residues indicate the positions of the unfolding barriers of the protein as illustrated in ﬁgure 6.14D. The other non-buried aromatic residues are in yellow. The structures which are excluded from unfolding, i.e. outside the cysteine handles (magenta) are indicated in gray. This includes two buried aromatic residues (blue).

of phenylalanine are zipping and packing together the antiparallel β-sheets of the ﬁbril which likely drives the misfolding and stabilizes the resulting structure. 444,445 The stability that arises due to the cohesion of the aromatic residues is similar to the hydrophobic collapse of globular proteins. However, the directed and sequential folding of Hbp has prohibited this scenario and instead led to the formation of secondary and tertiary structures, which in this case is the β-helical backbone. This indicates that the cohesive or attractive force between the stacked aromatic residues drives the folding of the polypeptide into β-helices, which in turn initiates the translocation of the whole passenger protein across the outer membrane. A high-throughput, scanning alanine mutagenesis study on the folding of the P22 tailspike by Simskovsky, et al 446 has shown that the buried aromatic core of the helical loops is critical for folding and that the stacking of these hydrophobic and aromatic residues drives the processive folding mechanism of the β-helices. A study by Soprova, et al 398 on Hbp has also shown that the mutation at W1015A (indicated in blue in ﬁgure 6.16) ac- tually results to a secretion block of Hbp. These studies and our experimental and calculation results strongly suggest that the stacking of the aromatic residues directs the processive folding of Hbp which then drives the transport of the whole passenger protein out of the outer membrane.

 6 Folding-driven secretion mechanism of autotransporters

6.3.5 Tracing the unfolding landscape of Hbp

To illustrate how the L805 truncate of Hbp stochastically follows the different barriers during unfolding, we created a map of the unfolding energy landscape (Figure 6.17). It shows the different barriers and the probability of the different pathways the protein can traverse during unfolding. The center of the circle represents the folded configuration of the protein while the unfolded configuration is represented by the outer diameter of the circle. The chart shows that from a folded state, the protein can unfold gradually in one step with 20% probability or in a sequential multi-step and multi-route unfolding pathways, e.g. in 2-, 3- and more-step pathways with probabilities of 42%, 25% and 15%, respectively. Figure 6.17A also illustrates the positions of the unfolding barriers (arcs) in relation to the unfolding axis and the probabilities that the protein encounters these barriers, e.g. at 42 nm with 40% probability, and almost equal probabilities at 30% the barriers at 15 nm, 24 nm, 32 nm and 48 nm. The unfolding forces for most of the barriers are about 80 pN (median) for loading speeds of 100 ∼500 nm/s as shown in figure 6.12C, except for those corresponding to the last two C-terminal β-helices which record 110 pN and 131 pN, respectively. A 3D representation of the unfolding pathways of L805 is created by approximating a conical energy landscape between the folded/unfolded conformations and Gaussian profile minima for the unfolding barriers, as shown in figure 6.17B&C. This illustrates the increase in the total energy of the protein as it is unfolded due to entropy and the presence of stable minima when an unfolding barrier is encountered. The minimum energy of the unfolding barriers are approximated equal as the

Figure 6.17 Unfolding routes of Hbp L805. A) Probability chart of the routes and pathways that Hbp can traverse during unfolding. The radius represents the length of L805: the center (up to 4 nm) and the outer diameter (57 nm) represent its folded and unfolded conformations, respectively. B,C) Unfolding energy landscape model (top and side views). The energy from unfolded to folded state is approximated by a cone and the unfolding/folding barriers by Gaussian minima. The radial lines represent the diﬀerent routes of unfolding.

 Results and Discussion unfolding forces obtained are similar as shown in ﬁgure 6.12B. This representation shows the approximate unfolding landscape of Hbp, where the radial lines represent the diﬀerent routes of unfolding and the angular widths represent the probabilities.

6.3.6 Hbp unfolds in domains

To recreate the actual unfolding of the wild type Hbp, the unfolding pathways of the longer constructs A558 and N256 were also investigated. As shown in figure 6.3, A558 is composed of the L805 and an additional eight β-helices towards the N-terminal region. The additional part is illustrated in more detail in figure 6.18A. Similarly to the L805, aromatic residues are buried inside the helical backbone, although not as regularly stacked as L805. A close inspection of the structure shows that the additional eight β-helices are connected to the rest of the L805 structure through a long linear linker. Furthermore, the top to bottom view of the structure shows that the first two additional β-helices are composed of three short strands instead of two (long and short) and the axis of these helices is gradually rotated up to about 60◦ relative to the β-helical backbone of L805. The linear linker is probably flexible and may present a weak spot in the protein, while the change in structure of the β-helical backbone may indicate a different unfolding characteristics. Figure 6.19A shows the Lc distributions of the unfolded fragments of A558 and the positions of the aromatic residues. Remarkably, the distribution for the last unfolding events closely resembles that of the L805 showing four prominent peaks which we can attribute to the unfolding of the four β-helices near the C-terminal. These results suggest that the L805 construct can be considered an independent unfolding domain that represents the unfolding-resistant part of the protein. Similar observations are found for the larger N256 truncate, which additionally comprises the chitinase b-like domain and the less-ordered part of the β-helical backbone which are mostly composed of two short β-strands. It also contains aromatic residues which are buried in the β-helical core but none is stacked on top of each other as illustrated in figure 6.18B. The Lc distribution for the last events shows unfolding pathways which are characteristic of the L805 truncate while the intermediate distribution shows a peak at the position of the two stacked residues F716 and F734. We note that many of the unfolding peaks seem to correspond to the positions of the buried aromatic residues. The calculations (refer to figures 6.23 and 6.24 in the Appendix) suggest sequential folding of the A558 but not the N-terminal region of N256. However, given that the first events experimentally observed in these long structures are short fragments relative to the protein truncate’s length, it is difficult to confirm experimentally that the unfolding is sequential despite the suggestive data. Nevertheless, the character-

 6 Folding-driven secretion mechanism of autotransporters

A B

F716 F734

linker linker

A558

Figure 6.18 A) Top-down and upright views of A558 truncate. The rotation of the additional eight β-helices relative to the L805 helices (gray) are indicated by the broken lines. Aromatic residues F716 and F734 are the only ones stacked on top of each other. The peripheral sections are indicated in green for emphasis. B) Top-down and upright views of N256 truncate (A558 in gray). None of the buried aromatic residues are stacked on top of each other. The additional β-loops are made of two short strands instead of long/short as in the L805 and A558 truncates. The chitinase b-like domain (green) has aromatic residues which are closely-packed between the β-strands (purple).

Figure 6.19 Unfolding pathways of A) A558 and B) N256. The plots are the Lc distributions of all, ﬁrst, intermediate and last unfolding events; and the positions of the aromatic residues and other structures of the protein, such as α helices, β strands and the chitinase b-like domain. The distributions of the last events for both truncates resemble that of the L805. The intermediate distributions show a clear peak at the position where the stacked aromatic residues F716 and F734 are located (?) and just before the chitinase domain starts (+).

 Results and Discussion

istics of the unfolding pathways (last events) and correspondence of the Lc peaks with the positions of the buried aromatic residues could suggest that Hbp unfolds sequentially in domains starting from the weaker structures in the N-terminal region to the structurally-stronger β-helical structures in the C-terminal region. The structural stability of the structure close to the C-terminal seemed to be held together and maintained by the interaction between the buried and stacked aromatic residues of the β-helical backbone. With this assumption, we can conversely hypothesize that the folding of the whole protein is sequential from C → N direction. The folding is initiated by the folding of the β-strands; and the attractive interaction of the aromatic residues directs the folding to a stable β-helical configuration. This processive folding in turn drives the transport of the rest of the protein across the outer membrane. The insertion of the first few residues of the C-terminal region through the β-barrel translocation domain might be initiated and assisted by the Bam complex. Once the first β-strands are outside the OM, the folding of the β-helical backbone becomes more energetically-favorable due to the π-stacking of the aromatic rings which forms a strong attractive force between the aromatic residues. When the first four or five β-helices are out of the OM, maintaining the entropic energy of staying inside the periplasm is now disadvantageous and costly for the rest of the protein. Thus, it becomes more favorable for the rest of the protein to unfold (if they are partly-folded in the periplasm) and form a linear polypeptide and be pulled out through the β- barrel by the folding of the β-helices outside the OM. Once the tipping of the balance of energy (to go out rather than to remain inside the periplasm) is accomplished by the folding of the first β-helices in the C-terminal, there is no more reason to create a strong β-helical structure. This could be the reason that fewer aromatic residues are stacked on top of each other beyond the L805 structure, but there are still a lot of buried aromatic residues to form the long and stable β-helical backbone.

 6 Folding-driven secretion mechanism of autotransporters

6.4 Conclusion

AFM imaging of Hbp wild type and truncates enable us to probe the rigid and flexible regions in the protein structure. The images indicate that the the C-terminal region which comprise the regularly-stacked β-helical backbone is quite stable and rigid, while the N-terminal region comprising the protease domain is hinged and flexible. The AFM unfolding experiments unravelled the initiation of the complex 1 unfolding pathways of Hbp and revealed that at least the first ⁄3 of the protein unfolds its β-helical structure in a sequential fashion. The unfolding landscape of the Hbp truncates show multi-step and multi-route unfolding pathways along fixed unfolding barriers. A comparison of the experimental results and a simple calculation of unfolding parameter based on the number of contacts and hydrogen bonds broken during unfolding show that the C-terminal region is resistant to unfolding and these unfolding barriers are marked by the presence of buried aromatic residues which are stacked on top of each other in the β-helical ladder. This indicates that the cohesive interaction between these aromatic residues drives a processive folding of the β-helical backbone and initiates the translocation of the entire protein across the outer membrane.

 Conclusion

Appendix

Figure 6.20 Lc distribution before and after translation. A) The groupings found in the Lc distribution (black) are Gaussian-fitted (broken lines). The peaks of the fits are used as the Lc’s of the model curve. Each Lc is approximated by a normalized Gaussian curve with a width of 4 nm. The Lc’s of each force-distance curves are also approximated with normalized Gaussian curves with widths of 4 nm and compared with the model curve. They are translated (positive and negative) with respect to this model curve and the cross-correlation between the translated and model curves (simple multiplication) are computed. Each force-distance curve is finally translated by the translation value which corresponds to the maximum cross-correlation. This process is repeated for all force-distance curves. The Lc distribution of the translated force-distance curves is shown in red. B) The distribution of the translation values applied to all force-distance curves. The results show that minor translations of the curves (± 1 nm) are enough to clean-up the distribution.

Figure 6.21 Maximum unfolding forces for diﬀerent pulling speeds. The distributions show that the forces are quite similar for loading speeds of 100-500 nm/s.

 6 Folding-driven secretion mechanism of autotransporters

Figure 6.22 Calculation of the energy in unfolding a fragment based on the number of contacts and hydrogen bonds broken, and search for the sequence yielding the lowest energy. A,B) The energy plots in unfolding a fragment with length Lc and the possible sequences. Shown are the start and end sequences of the amino acids comprising the fragment. C) The sequence yielding the minimum unfolding energy for the corresponding Lc. The plot demonstrates that the protein favors sequential unfolding from the N-terminal (position 805). D) Energy and derivative of the energy of the fragments indicated in C. The peaks correspond to breaking large chunks of contacts and hydrogen bonds to unfold the protein structures labelled by the sequence numbers. The end sequence of these fragments is indicated by the amino acid in yellow.

 Conclusion

Figure 6.23 Calculation of the energy in unfolding Hbp A558 construct. A) The sequence yielding the minimum unfolding energy. B) Energy and derivative of the energy of the fragments indicated in A. The inset shows a section which resemble that of L805.

Figure 6.24 Calculation of the energy in unfolding Hbp N256 construct. A) The sequence yielding the minimum unfolding energy. The plot demonstrate non-sequential unfolding speciﬁcally the fragments with lengths below 110 nm. B) Energy and derivative of the energy of the fragments indicated in A. The inset shows a section which resemble that of A558.

 Bibliography

[1] G. Bao, “Mechanics of biomolecules,” Journal of the Mechanics and Physics of Solids, vol. 50, no. 11, pp. 2237–2274, 2002.

[2] J. Eyckmans, T. Boudou, X. Yu, and C. S. Chen, “A Hitchhiker’s Guide to Mechanobi- ology,” Developmental Cell, vol. 21, no. 1, SI, pp. 35–47, 2011.

[3] Y. Sun, C. S. Chen, and J. Fu, “Forcing Stem Cells to Behave: A Biophysical Perspec- tive of the Cellular Microenvironment,” in Annual Review of Biophysics (Rees, DC, ed.), vol. 41 of Annual Review of Biophysics, pp. 519–542, 2012.

[4] T. Mammoto and D. E. Ingber, “Mechanical control of tissue and organ development,” Development, vol. 137, no. 9, pp. 1407–1420, 2010.

[5] D. Ingber, “Mechanobiology and diseases of mechanotransduction,” Annals of Medicine, vol. 35, no. 8, pp. 564–577, 2003.

[6] B. Huang, M. Bates, and X. Zhuang, “Super-Resolution Fluorescence Microscopy,” Annual Review of Biochemistry, vol. 78, pp. 993–1016, 2009.

[7] R. Schmidt, C. A. Wurm, S. Jakobs, J. Engelhardt, A. Egner, and S. W. Hell, “Spherical nanosized focal spot unravels the interior of cells,” Nature Methods, vol. 5, no. 6, pp. 539–544, 2008.

[8] M. Bates, B. Huang, G. T. Dempsey, and X. Zhuang, “Multicolor super-resolution imaging with photo-switchable ﬂuorescent probes,” Science, vol. 317, no. 5845, pp. 1749–1753, 2007.

[9] N. Crampton, W. A. Bonass, J. Kirkham, C. Rivetti, and N. H. Thomson, “Collision events between RNA polymerases in convergent transcription studied by atomic force microscopy,” Nucleic Acids Research, vol. 34, no. 19, pp. 5416–5425, 2006.

[10] K. C. Neuman and A. Nagy, “Single-molecule force spectroscopy: optical tweezers, magnetic tweezers and atomic force microscopy,” Nature Methods, vol. 5, no. 6, pp. 491– 505, 2008. BIBLIOGRAPHY

[11] A. Ashkin, J. M. Dziedzic, J. E. Bjorkholm, and S. Chu, “Observation of a single- beam gradient force optical trap for dielectric particles,” Optics Letters, vol. 11, no. 5, pp. 288–290, 1986.

[12] K. Berg-Sorensen and H. Flyvbjerg, “Power spectrum analysis for optical tweezers,” Review of Scientiﬁc Instruments, vol. 75, no. 3, pp. 594–612, 2004.

[13] K. Neuman and S. Block, “Optical trapping,” Review of Scientiﬁc Instruments, vol. 75, no. 9, pp. 2787–2809, 2004.

[14] C. Gosse and V. Croquette, “Magnetic tweezers: Micromanipulation and force measurement at the molecular level,” Biophysical Journal, vol. 82, no. 6, pp. 3314–3329, 2002.

[15] T. Strick, V. Croquette, and D. Bensimon, “Single-molecule analysis of DNA uncoiling by a type II topoisomerase,” Nature, vol. 404, no. 6780, pp. 901–904, 2000.

[16] H. Itoh, A. Takahashi, K. Adachi, H. Noji, R. Yasuda, M. Yoshida, and K. Ki-

nosita, “Mechanically driven ATP synthesis by F1-ATPase,” Nature, vol. 427, no. 6973, pp. 465–468, 2004.

[17] G. Binnig, C. F. Quate, and C. Gerber, “Atomic force microscope,” Physical Review Letters, vol. 56, no. 9, pp. 930–933, 1986.

[18] M. Baclayon, G. J. L. Wuite, and W. H. Roos, “Imaging and manipulation of single viruses by atomic force microscopy,” Soft Matter, vol. 6, no. 21, pp. 5273–5285, 2010.

[19] A. Borgia, P. M. Williams, and J. Clarke, “Single-molecule studies of protein folding,” Annual Review of Biochemistry, vol. 77, pp. 101–125, 2008.

[20] F. Ritort, “Single-molecule experiments in biological physics: methods and applications,” Journal of Physics - Condensed Matter, vol. 18, no. 32, pp. R531–R583, 2006. Symposium on Low Friction Coatings on Glass and Ceramics, Baltimore, MD, APR, 2005.

[21] E. W. Frey, A. A. Gooding, S. Wijeratne, and C.-H. Kiang, “Understanding the physics of DNA using nanoscale single-molecule manipulation,” Frontiers of Physics, vol. 7, no. 5, pp. 576–581, 2012.

[22] C. P. Calderon, W.-H. Chen, K.-J. Lin, N. C. Harris, and C.-H. Kiang, “Quantifying DNA melting transitions using single-molecule force spectroscopy,” Journal of Physics - Condensed Matter, vol. 21, no. 3, 2009.

[23] P. Gross, N. Laurens, L. B. Oddershede, U. Bockelmann, E. J. G. Peterman, and G. J. L. Wuite, “Quantifying how DNA stretches, melts and changes twist under tension,” Nature Physics, vol. 7, no. 9, pp. 731–736, 2011.

 BIBLIOGRAPHY

[24] E. Botello, N. C. Harris, J. Sargent, W.-H. Chen, K.-J. Lin, and C.-H. Kiang, “Temper- ature and Chemical Denaturant Dependence of Forced Unfolding of Titin 127,” Journal of Physical Chemistry B, vol. 113, no. 31, pp. 10845–10848, 2009.

[25] R. Dahm, “Discovering DNA: Friedrich Miescher and the early years of nucleic acid research,” Human Genetics, vol. 122, no. 6, pp. 565–581, 2008.

[26] J. Watson and F. Crick, “Molecular Structure of Nucleic Acids - A Structure for De- oxyribose Nucleic Acid,” Nature, vol. 171, no. 4356, pp. 737–738, 1953.

[27] C. Bustamante, J. Marko, E. Siggia, and S. Smith, “Entropic Elasticity of Lambda- Phage DNA,” Science, vol. 265, no. 5178, pp. 1599–1600, 1994.

[28] J. Marko and E. Siggia, “Stretching DNA,” Macromolecules, vol. 28, no. 26, pp. 8759– 8770, 1995.

[29] T. Odijk, “Stiﬀ Chains and Filaments under Tension,” Macromolecules, vol. 28, no. 20, pp. 7016–7018, 1995.

[30] C. M. Franz and P. H. Puech, “Atomic Force Microscopy: A Versatile Tool for Studying Cell Morphology, Adhesion and Mechanics,” Cellular and Molecular Bioengineering, vol. 1, no. 4, pp. 289–300, 2008.

[31] I. A. T. Schaap, C. Carrasco, P. J. de Pablo, and C. F. Schmidt, “Kinesin Walks the Line: Single Motors Observed by Atomic Force Microscopy,” Biophysical Journal, vol. 100, no. 10, pp. 2450–2456, 2011.

[32] S. Safran, N. Gov, A. Nicolas, U. Schwarz, and T. Tlusty, “Physics of cell elasticity, shape and adhesion,” Physica A - Statistical Mechanics and its Applications, vol. 352, no. 1, pp. 171–201, 2005.

[33] M. L. Gardel, K. E. Kasza, C. P. Brangwynne, J. Liu, and D. A. Weitz, “Mechanical Response of Cytoskeletal Networks,” in Biophysical Tools for Biologists, Vol. 2: In vivo techniques (Correia, JJ and Detrich, HW, ed.), vol. 89 of Methods in Cell Biology, pp. 487+, 2008.

[34] W. Roos, A. Roth, J. Konle, H. Presting, E. Sackmann, and J. Spatz, “Freely suspended actin cortex models on arrays of microfabricated pillars,” ChemPhysChem, vol. 4, no. 8, pp. 872–877, 2003.

[35] L. Trichet, J. Le Digabel, R. J. Hawkins, R. K. Vedula, M. Gupta, C. Ribrault, P. Hersen, R. Voituriez, and B. Ladoux, “Evidence of a large-scale mechanosensing mechanism for cellular adaptation to substrate stiﬀness,” Proceedings of the National Academy of Sciences USA, vol. 109, no. 18, pp. 6933–6938, 2012.

 BIBLIOGRAPHY

[36] D. Discher and P. Carl, “New insights into red cell network structure, elasticity, and spectrin unfolding - A current review,” Cellular & Molecular Biology Letters, vol. 6, no. 3, pp. 593–606, 2001.

[37] P. J. de Pablo, I. A. T. Schaap, F. C. MacKintosh, and C. F. Schmidt, “Deformation and collapse of microtubules on the nanometer scale,” Physical Review Letters, vol. 91, no. 9, p. 4, 2003.

[38] S. Nawaz, P. Sanchez, K. Bodensiek, S. Li, M. Simons, and I. A. T. Schaap, “Cell Visco-Elasticity Measured with AFM and Optical Trapping at Sub-Micrometer Defor- mations,” PLOS One, vol. 7, no. 9, 2012.

[39] N. E. Kurland, Z. Drira, and V. K. Yadavalli, “Measurement of nanomechanical properties of biomolecules using atomic force microscopy,” Micron, vol. 43, no. 2-3, pp. 116– 128, 2012.

[40] K. E. Kasza, A. C. Rowat, J. Liu, T. E. Angelini, C. P. Brangwynne, G. H. Koenderink, and D. A. Weitz, “The cell as a material,” Current Opinion in Cell Biology, vol. 19, no. 1, pp. 101–107, 2007.

[41] Q. Wen and P. A. Janmey, “Polymer physics of the cytoskeleton,” Current Opinion in Solid State & Material Science, vol. 15, no. 5, pp. 177–182, 2011.

[42] C. P. Broedersz, C. Storm, and F. C. MacKintosh, “Nonlinear elasticity of composite networks of stiﬀ biopolymers with ﬂexible linkers,” Physical Review Letters, vol. 101, no. 11, 2008.

[43] P. Bursac, G. Lenormand, B. Fabry, M. Oliver, D. Weitz, V. Viasnoﬀ, J. Butler, and J. Fredberg, “Cytoskeletal remodelling and slow dynamics in the living cell,” Nature Materials, vol. 4, no. 7, pp. 557–561, 2005.

[44] W. H. Roos, R. Bruinsma, and G. J. L. Wuite, “Physical virology,” Nature Physics, vol. 6, no. 10, pp. 733–743, 2010.

[45] M. Stuart and R. Nagel, “Sickle-cell disease,” Lancet, vol. 364, no. 9442, pp. 1343–1360, 2004.

[46] Q. S. Li, G. Y. H. Lee, C. N. Ong, and C. T. Lim, “Afm indentation study of breast cancer cells,” Biochemical and Biophysical Research Communications, vol. 374, no. 4, pp. 609–613, 2008.

[47] N. Kol, Y. Shi, M. Tsvitov, D. Barlam, R. Z. Shneck, M. S. Kay, and I. Rousso, “A stiﬀness switch in Human Immunodeﬁciency Virus,” Biophysical Journal, vol. 92, no. 5, pp. 1777–1783, 2007.

 BIBLIOGRAPHY

[48] A. Ikai, “A Review on: Atomic Force Microscopy Applied to Nano-mechanics of the Cell,” in Nano/Micro Biotechnology (Endo, I and Nagamune, T, ed.), vol. 119 of Advances in Biochemical Engineering-Biotechnology, pp. 47–61, 2010.

[49] L. D. Landau and E. M. Lifshitz, Theory of Elasticity. Oxford: Butterworth- Heinemann, third ed., 1986.

[50] W. S. Klug, R. F. Bruinsma, J. P. Michel, C. M. Knobler, I. L. Ivanovska, C. F. Schmidt, and G. J. L. Wuite, “Failure of viral shells,” Physical Review Letters, vol. 97, no. 22, p. 228101, 2006.

[51] W. H. Roos, M. M. Gibbons, A. Arkhipov, C. Uetrecht, N. R. Watts, P. T. Wingﬁeld, A. C. Steven, A. J. R. Heck, K. Schulten, W. S. Klug, and G. J. L. Wuite, “Squeezing protein shells: how continuum elastic models, molecular dynamics simulations and experiments coalesce at the nanoscale,” Biophysical Journal, vol. in press, 2010.

[52] M. Baclayon, W. H. Roos, and G. J. L. Wuite, “Sampling Protein Form and Function with the Atomic Force Microscope,” Molecular & Cellular Proteomics, vol. 9, no. 8, pp. 1678–1688, 2010.

[53] A. J. R. Heck, “Native mass spectrometry: a bridge between interactomics and structural biology,” Nature Methods, vol. 5, no. 11, pp. 927–933, 2008.

[54] S. Scheuring and J. N. Sturgis, “Chromatic adaptation of photosynthetic membranes,” Science, vol. 309, no. 5733, pp. 484–487, 2005.

[55] P. Fechner, T. Boudier, S. Mangenot, S. Jaroslawski, J. N. Sturgis, and S. Scheur- ing, “Structural information, resolution, and noise in high-resolution atomic force microscopy topographs,” Biophysical Journal, vol. 96, no. 9, pp. 3822–3831, 2009.

[56] E. Henderson, “Imaging and nanodissection of individual supercoiled plasmids by atomic force microscopy,” Nucleic Acids Research, vol. 20, no. 3, pp. 445–447, 1992.

[57] A. Engel and D. J. Muller, “Observing single biomolecules at work with the atomic force microscope,” Nature Structural Biology, vol. 7, no. 9, pp. 715–718, 2000.

[58] S. Scheuring, D. Fotiadis, C. Moller, S. A. Muller, A. Engel, and D. J. Muller, “Single proteins observed by atomic force microscopy,” Single Molecules, vol. 2, no. 2, pp. 59– 67, 2001.

[59] J. P. Junker, F. Ziegler, and M. Rief, “Ligand-dependent equilibrium ﬂuctuations of single calmodulin molecules,” Science, vol. 323, no. 5914, pp. 633–637, 2009.

[60] G. M. King, A. R. Carter, A. B. Churnside, L. S. Eberle, and T. T. Perkins, “Ul- trastable atomic force microscopy: Atomic-scale stability and registration in ambient conditions,” Nano Letters, vol. 9, no. 4, pp. 1451–1456, 2009.

 BIBLIOGRAPHY

[61] C. L. Cheung, J. H. Hafner, and C. M. Lieber, “Carbon nanotube atomic force microscopy tips: Direct growth by chemical vapor deposition and application to high- resolution imaging,” Proceedings of the National Academy of Sciences of the United States of America, vol. 97, no. 8, pp. 3809–3813, 2000.

[62] J. H. Hafner, C. L. Cheung, A. T. Woolley, and C. M. Lieber, “Structural and functional imaging with carbon nanotube AFM probes,” Progress in Biophysics & Molecular Biology, vol. 77, no. 1, pp. 73–110, 2001.

[63] P. Hinterdorfer, W. Baumgartner, H. J. Gruber, K. Schilcher, and H. Schindler, “Detec- tion and localization of individual antibody-antigen recognition events by atomic force microscopy,” Proceedings of the National Academy of Sciences of the United States of America, vol. 93, no. 8, pp. 3477–3481, 1996.

[64] S. S. Wong, E. Joselevich, A. T. Woolley, C. L. Cheung, and C. M. Lieber, “Covalently functionalized nanotubes as nanometre-sized probes in chemistry and biology,” Nature, vol. 394, no. 6688, pp. 52–55, 1998.

[65] P. Hinterdorfer and Y. F. Dufrene, “Detection and localization of single molecular recognition events using atomic force microscopy,” Nature Methods, vol. 3, no. 5, pp. 347–355, 2006.

[66] A. Ebner, P. Hinterdorfer, and H. J. Gruber, “Comparison of diﬀerent aminofunction- alization strategies for attachment of single antibodies to AFM cantilevers,” Ultrami- croscopy, vol. 107, no. 10-11, pp. 922–927, 2007.

[67] A. Ebner, L. Wildling, A. S. M. Kamruzzahan, C. Rankl, J. Wruss, C. D. Hahn, M. Holzl, R. Zhu, F. Kienberger, D. Blaas, P. Hinterdorfer, and H. J. Gruber, “A new, simple method for linking of antibodies to atomic force microscopy tips,” Bioconjugate Chemistry, vol. 18, no. 4, pp. 1176–1184, 2007.

[68] A. Ebner, L. Wildling, R. Zhu, C. Rankl, T. Haselgrubler, P. Hinterdorfer, and H. J. Gruber, Functionalization of probe tips and supports for single-molecule recognition force Microscopy, vol. 285 of Topics in Current Chemistry, pp. 29–76. Berlin: Springer- Verlag Berlin, 2008.

[69] A. Hards, C. Q. Zhou, M. Seitz, C. Brauchle, and A. Zumbusch, “Simultaneous AFM manipulation and ﬂuorescence imaging of single DNA strands,” Chemphyschem, vol. 6, no. 3, pp. 534–540, 2005.

[70] R. Kassies, K. O. Van der Werf, A. Lenferink, C. N. Hunter, J. D. Olsen, V. Sub- ramaniam, and C. Otto, “Combined AFM and confocal ﬂuorescence microscope for applications in bio-nanotechnology,” Journal of Microscopy-Oxford, vol. 217, pp. 109– 116, 2005.

 BIBLIOGRAPHY

[71] S. K. Kufer, E. M. Puchner, H. Gumpp, T. Liedl, and H. E. Gaub, “Single-molecule cut-and-paste surface assembly,” Science, vol. 319, no. 5863, pp. 594–596, 2008.

[72] T. Sulchek, R. Hsieh, J. D. Adams, S. C. Minne, C. F. Quate, and D. M. Adder- ton, “High-speed atomic force microscopy in liquid,” Review of Scientiﬁc Instruments, vol. 71, no. 5, pp. 2097–2099, 2000.

[73] T. Ando, N. Kodera, E. Takai, D. Maruyama, K. Saito, and A. Toda, “A high- speed atomic force microscope for studying biological macromolecules,” Proceedings of the National Academy of Sciences of the United States of America, vol. 98, no. 22, pp. 12468–12472, 2001.

[74] P. K. Hansma, G. Schitter, G. E. Fantner, and C. Prater, “Applied physics - high-speed atomic force microscopy,” Science, vol. 314, no. 5799, pp. 601–602, 2006.

[75] I. Casuso, N. Kodera, C. Le Grimellec, T. Ando, and S. Scheuring, “Contact-mode high-resolution high-speed atomic force microscopy movies of the purple membrane,” Biophysical Journal, vol. 97, no. 5, pp. 1354–1361, 2009.

[76] G. E. Fantner, R. J. Barbero, D. S. Gray, and A. M. Belcher, “Kinetics of antimicrobial peptide activity measured on individual bacterial cells using high-speed atomic force microscopy,” Nature Nanotechnology, vol. 5, no. 4, pp. 280–285, 2010.

[77] Y. F. Dufrene, “Atomic force microscopy: A powerful molecular toolkit in nanopro- teomics,” Proteomics, vol. 9, no. 24, pp. 5400–5405, 2009.

[78] X. B. Yu, D. K. Xu, and Q. Cheng, “Label-free detection methods for protein microar- rays,” Proteomics, vol. 6, no. 20, pp. 5493–5503, 2006.

[79] A. I. Archakov, Y. D. Ivanov, A. V. Lisitsa, and V. G. Zgoda, “AFM ﬁshing nanotechnology is the way to reverse the Avogadro number in proteomics,” Proteomics, vol. 7, no. 1, pp. 4–9, 2007.

[80] M. Rief and H. Grubmuller, “Force spectroscopy of single biomolecules,” Chemphyschem, vol. 3, no. 3, pp. 255–261, 2002.

[81] G. C. Zeng, J. B. Chen, L. Y. Zhong, R. Wang, L. F. Jiang, J. Y. Cai, L. Yan, D. Huang, C. Y. Chen, and Z. W. Chen, “NSOM- and AFM-based nanotechnology elucidates nano-structural and atomic-force features of a Y-pestis V immunogen-containing particle vaccine capable of eliciting robust response,” Proteomics, vol. 9, no. 6, pp. 1538– 1547, 2009.

[82] S. Kasas and G. Dietler, “Probing nanomechanical properties from biomolecules to living cells,” Pﬂugers Archiv-European Journal of Physiology, vol. 456, no. 1, pp. 13– 27, 2008.

 BIBLIOGRAPHY

[83] W. H. Roos and G. J. L. Wuite, “Nanoindentation studies reveal material properties of viruses,” Advanced Materials, vol. 21, pp. 1187–1192, 2009.

[84] G. Binnig, C. Gerber, E. Stoll, T. R. Albrecht, and C. F. Quate, “Atomic resolution with atomic force microscope,” Europhysics Letters, vol. 3, no. 12, pp. 1281–1286, 1987.

[85] F. J. Giessibl and G. Binnig, “Investigation of the (001) cleavage plane of potassium- bromide with an atomic force microscope at 4.2-k in ultra-high vacuum,” Ultrami- croscopy, vol. 42, pp. 281–289, 1992.

[86] J. Zlatanova and S. H. Leuba, “Stretching and imaging single DNA molecules and chromatin,” Journal of Muscle Research and Cell Motility, vol. 23, no. 5-6, pp. 377– 395, 2002.

[87] P. Parot, Y. F. Dufrene, P. Hinterdorfer, C. Le Grimellec, D. Navajas, J. L. Pellequer, and S. Scheuring, “Past, present and future of atomic force microscopy in life sciences and medicine,” Journal of Molecular Recognition, vol. 20, pp. 418–431, 2007.

[88] H. G. Hansma and J. H. Hoh, “Biomolecular imaging with the atomic-force microscope,” Annual Review Of Biophysics And Biomolecular Structure, vol. 23, pp. 115–139, 1994.

[89] F. J. Giessibl, “Advances in atomic force microscopy,” Reviews of Modern Physics, vol. 75, pp. 949–983, 2003.

[90] N. Jalili and K. Laxminarayana, “A review of atomic force microscopy imaging systems: application to molecular metrology and biological sciences,” Mechatronics, vol. 14, no. 8, pp. 907–945, 2004.

[91] R. Garcia and R. Perez, “Dynamic atomic force microscopy methods,” Surface Science Reports, vol. 47, no. 6-8, pp. 197–301, 2002.

[92] P. C. Braga and D. Ricci, Atomic force microscopy : biomedical methods and applications. Methods in molecular biology, Totowa, N.J.: Humana Press, 2004.

[93] V. J. Morris, A. R. Kirby, and A. P. Gunning, Atomic force microscopy for biologists. London: Imperial College Press, 2nd ed., 2010.

[94] P. J. de Pablo, J. Colchero, J. Gomez-Herrero, and A. M. Baro, “Jumping mode scanning force microscopy,” Applied Physics Letters, vol. 73, no. 22, pp. 3300–3302, 1998.

[95] Q. Zhong, D. Inniss, K. Kjoller, and V. B. Elings, “Fractured polymer silica ﬁber surface studied by tapping mode atomic force microscopy,” Surface Science, vol. 290, no. 1-2, pp. L688–L692, 1993.

[96] B. Cappella and G. Dietler, “Force-distance curves by atomic force microscopy,” Surface Science Reports, vol. 34, no. 1-3, pp. 1–104, 1999.

 BIBLIOGRAPHY

[97] H. J. Butt, B. Cappella, and M. Kappl, “Force measurements with the atomic force microscope: Technique, interpretation and applications,” Surface Science Reports, vol. 59, no. 1-6, pp. 1–152, 2005.

[98] A. Chilkoti, T. Boland, B. D. Ratner, and P. S. Stayton, “The relationship between ligand-binding thermodynamics and protein-ligand interaction forces measured by atomic force microscopy,” Biophysical Journal, vol. 69, no. 5, pp. 2125–2130, 1995.

[99] H. Takano, J. R. Kenseth, S. S. Wong, J. C. O’Brien, and M. D. Porter, “Chemical and biochemical analysis using scanning force microscopy,” Chemical Reviews, vol. 99, no. 10, pp. 2845–2890, 1999.

[100] D. Leckband, “Measuring the forces that control protein interactions,” Annual Review of Biophysics and Biomolecular Structure, vol. 29, pp. 1–26, 2000.

[101] J. Zlatanova, S. M. Lindsay, and S. H. Leuba, “Single molecule force spectroscopy in biology using the atomic force microscope,” Progress in Biophysics & Molecular Biology, vol. 74, no. 1-2, pp. 37–61, 2000.

[102] A. F. Oberhauser, P. K. Hansma, M. Carrion-Vazquez, and J. M. Fernandez, “Stepwise unfolding of titin under force-clamp atomic force microscopy,” Proceedings of the Na- tional Academy of Sciences of the United States of America, vol. 98, no. 2, pp. 468–472, 2001.

[103] W. A. Linke and A. Grutzner, “Pulling single molecules of titin by AFM - recent advances and physiological implications,” Pﬂugers Archiv-European Journal of Physi- ology, vol. 456, no. 1, pp. 101–115, 2008.

[104] J. M. Fernandez and H. B. Li, “Force-clamp spectroscopy monitors the folding trajec- tory of a single protein,” Science, vol. 303, no. 5664, pp. 1674–1678, 2004.

[105] M. Schlierf, H. B. Li, and J. M. Fernandez, “The unfolding kinetics of ubiquitin captured with single-molecule force-clamp techniques,” Proceedings of the National Academy of Sciences of the United States of America, vol. 101, no. 19, pp. 7299–7304, 2004.

[106] T. E. Fisher, P. E. Marszalek, and J. M. Fernandez, “Stretching single molecules into novel conformations using the atomic force microscope,” Nature Structural Biology, vol. 7, no. 9, pp. 719–724, 2000.

[107] F. Oesterhelt, D. Oesterhelt, M. Pfeiﬀer, A. Engel, H. E. Gaub, and D. J. Muller, “Unfolding pathways of individual bacteriorhodopsins,” Science, vol. 288, no. 5463, pp. 143–146, 2000.

[108] A. Kedrov, H. Janovjak, K. T. Sapra, and D. J. Muller, “Deciphering molecular interactions of native membrane proteins by single-molecule force spectroscopy,” Annual Review of Biophysics and Biomolecular Structure, vol. 36, pp. 233–260, 2007.

 BIBLIOGRAPHY

[109] A. Vinckier, P. Gervasoni, F. Zaugg, U. Ziegler, P. Lindner, P. Groscurth, A. Pluckthun, and G. Semenza, “Atomic force microscopy detects changes in the interaction forces between GroEL and substrate proteins,” Biophysical Journal, vol. 74, no. 6, pp. 3256– 3263, 1998.

[110] O. H. Willemsen, M. M. E. Snel, A. Cambi, J. Greve, B. G. De Grooth, and C. G. Figdor, “Biomolecular interactions measured by atomic force microscopy,” Biophysical Journal, vol. 79, no. 6, pp. 3267–3281, 2000.

[111] T. J. Senden and W. A. Ducker, “Experimental determination of spring constants in atomic force microscopy,” Langmuir, vol. 10, no. 4, pp. 1003–1004, 1994.

[112] C. A. Cliﬀord and M. P. Seah, “The determination of atomic force microscope cantilever spring constants via dimensional methods for nanomechanical analysis,” Nanotechnol- ogy, vol. 16, no. 9, pp. 1666–1680, 2005.

[113] T. R. Albrecht, S. Akamine, T. E. Carver, and C. F. Quate, “Microfabrication of cantilever styli for the atomic force microscope,” Journal of Vacuum Science & Technology A -Vacuum Surfaces and Films, vol. 8, no. 4, pp. 3386–3396, 1990.

[114] G. Y. Chen, R. J. Warmack, T. Thundat, D. P. Allison, and A. Huang, “Resonance response of scanning force microscopy cantilevers,” Review of Scientiﬁc Instruments, vol. 65, no. 8, pp. 2532–2537, 1994.

[115] J. L. Hutter and J. Bechhoefer, “Calibration of atomic-force microscope tips,” Review of Scientiﬁc Instruments, vol. 64, no. 7, pp. 1868–1873, 1993.

[116] J. E. Sader, I. Larson, P. Mulvaney, and L. R. White, “Method for the calibration of atomic force microscope cantilevers,” Review of Scientiﬁc Instruments, vol. 66, no. 7, pp. 3789–3798, 1995.

[117] J. E. Sader, J. W. M. Chon, and P. Mulvaney, “Calibration of rectangular atomic force microscope cantilevers,” Review of Scientiﬁc Instruments, vol. 70, no. 10, pp. 3967– 3969, 1999.

[118] M. J. Allen, N. V. Hud, M. Balooch, R. J. Tench, W. J. Siekhaus, and R. Balhorn, “Tip-radius-induced artifacts in AFM images of protamine-complexed DNA ﬁbers,” Ultramicroscopy, vol. 42, pp. 1095–1100, 1992.

[119] T. Thundat, X. Y. Zheng, S. L. Sharp, D. P. Allison, R. J. Warmack, D. C. Joy, and T. L. Ferrell, “Calibration of atomic force microscope tips using biomolecules,” Scanning Microscopy, vol. 6, no. 4, pp. 903–910, 1992.

[120] B. V. V. Prasad, R. Rothnagel, X. Jiang, and M. K. Estes, “3-dimensional structure of baculovirus-expressed Norwalk virus capsids,” Journal of Virology, vol. 68, no. 8, pp. 5117–5125, 1994.

 BIBLIOGRAPHY

[121] B. V. V. Prasad, M. E. Hardy, T. Dokland, J. Bella, M. G. Rossmann, and M. K. Estes, “X-ray crystallographic structure of the Norwalk virus capsid,” Science, vol. 286, no. 5438, pp. 287–290, 1999.

[122] G. K. Shoemaker, E. van Duijn, S. E. Crawford, C. Uetrecht, M. Baclayon, W. H. Roos, G. J. L. Wuite, M. K. Estes, B. V. V. Prasad, and A. J. R. Heck, “Norwalk Virus Assembly and Stability Monitored by Mass Spectrometry,” Molecular & Cellular Proteomics, vol. 9, no. 8, pp. 1742–1751, 2010.

[123] D. F. Ogletree, R. W. Carpick, and M. Salmeron, “Calibration of frictional forces in atomic force microscopy,” Review of Scientiﬁc Instruments, vol. 67, no. 9, pp. 3298– 3306, 1996.

[124] C. Staii, D. W. Wood, and G. Scoles, “Veriﬁcation of biochemical activity for proteins nanografted on gold surfaces,” Journal of the American Chemical Society, vol. 130, no. 2, pp. 640–646, 2008.

[125] M. Gladnikoﬀ, E. Shimoni, N. S. Gov, and I. Rousso, “Retroviral assembly and budding occur through an actin-driven mechanism,” Biophysical Journal, vol. 97, no. 9, pp. 2419–2428, 2009.

[126] S. N. Magonov, V. Elings, and M. H. Whangbo, “Phase imaging and stiﬀness in tapping- mode atomic force microscopy,” Surface Science, vol. 375, no. 2-3, pp. L385–L391, 1997.

[127] J. Melcher, C. Carrasco, X. Xu, J. L. Carrascosa, J. Gomez-Herrero, P. J. de Pablo, and A. Raman, “Origins of phase contrast in the atomic force microscope in liquids,” Proceedings of the National Academy of Sciences of the United States of America, vol. 106, no. 33, pp. 13655–13660, 2009.

[128] D. L. Worcester, R. G. Miller, and P. J. Bryant, “Atomic force microscopy of purple membranes,” Journal of Microscopy-Oxford, vol. 152, pp. 817–821, 1988.

[129] Y. L. Lyubchenko, A. A. Gall, L. S. Shlyakhtenko, R. E. Harrington, B. L. Jacobs, P. I. Oden, and S. M. Lindsay, “Atomic force microscopy imaging of double-stranded DNA and RNA,” Journal of Biomolecular Structure & Dynamics, vol. 10, no. 3, pp. 589–606, 1992.

[130] H. G. Hansma, M. Bezanilla, F. Zenhausern, M. Adrian, and R. L. Sinsheimer, “Atomic force microscopy of DNA in aqueous-solutions,” Nucleic Acids Research, vol. 21, no. 3, pp. 505–512, 1993.

[131] S. Kasas, N. H. Thomson, B. L. Smith, P. K. Hansma, J. Miklossy, and H. G. Hansma, “Biological applications of the AFM: From single molecules to organs,” International Journal of Imaging Systems and Technology, vol. 8, no. 2, pp. 151–161, 1997.

 BIBLIOGRAPHY

[132] D. J. Muller, H. Janovjak, T. Lehto, L. Kuerschner, and K. Anderson, “Observing structure, function and assembly of single proteins by AFM,” Progress in Biophysics & Molecular Biology, vol. 79, no. 1-3, pp. 1–43, 2002.

[133] B. Drake, C. B. Prater, A. L. Weisenhorn, S. A. C. Gould, T. R. Albrecht, C. F. Quate, D. S. Cannell, H. G. Hansma, and P. K. Hansma, “Imaging crystals, polymers, and processes in water with the atomic force microscope,” Science, vol. 243, no. 4898, pp. 1586–1589, 1989.

[134] W. Haberle, J. K. H. Horber, F. Ohnesorge, D. P. E. Smith, and G. Binnig, “In situ investigations of single living cells infected by viruses,” Ultramicroscopy, vol. 42, pp. 1161–1167, 1992.

[135] Y. G. Kuznetsov, A. J. Malkin, R. W. Lucas, M. Plomp, and A. McPherson, “Imaging of viruses by atomic force microscopy,” Journal of General Virology, vol. 82, pp. 2025– 2034, 2001.

[136] M. Valle, J. M. Valpuesta, J. L. Carrascosa, J. Tamayo, and R. Garcia, “The interaction of DNA with bacteriophage phi 29 connector: A study by AFM and TEM,” Journal of Structural Biology, vol. 116, no. 3, pp. 390–398, 1996.

[137] D. J. Muller, A. Engel, J. L. Carrascosa, and M. Velez, “The bacteriophage phi 29 head-tail connector imaged at high resolution with the atomic force microscope in buﬀer solution,” Embo Journal, vol. 16, no. 10, pp. 2547–2553, 1997.

[138] W. H. Roos, K. Radtke, E. Kniesmeijer, H. Geertsema, B. Sodeik, and G. J. L. Wuite, “Scaﬀold expulsion and genome packaging trigger stabilization of Herpes Simplex virus capsids,” Proceedings of the National Academy of Sciences of the United States of America, vol. 106, no. 24, pp. 9673–9678, 2009.

[139] F. M. Ohnesorge, J. K. H. Horber, W. Haberle, C. P. Czerny, D. P. E. Smith, and G. Binnig, “AFM review study on pox viruses and living cells,” Biophysical Journal, vol. 73, no. 4, pp. 2183–2194, 1997.

[140] M. Gladnikoﬀ and I. Rousso, “Directly monitoring individual retrovirus budding events using atomic force microscopy,” Biophysical Journal, vol. 94, no. 1, pp. 320–326, 2008.

[141] W. F. Kolbe, D. F. Ogletree, and M. B. Salmeron, “Atomic force microscopy imaging of T4 bacteriophages on silicon substrates,” Ultramicroscopy, vol. 42, pp. 1113–1117, 1992.

[142] F. Kienberger, R. Zhu, R. Moser, D. Blaas, and P. Hinterdorfer, “Monitoring RNA release from human rhinovirus by dynamic force microscopy,” Journal of Virology, vol. 78, no. 7, pp. 3203–3209, 2004.

 BIBLIOGRAPHY

[143] Y. G. Kuznetsov, S. Daijogo, J. Zhou, B. L. Semler, and A. McPherson, “Atomic force microscopy analysis of icosahedral virus RNA,” Journal of Molecular Biology, vol. 347, no. 1, pp. 41–52, 2005.

[144] D. J. Muller, K. T. Sapra, S. Scheuring, A. Kedrov, P. L. Frederix, D. Fotiadis, and A. Engel, “Single-molecule studies of membrane proteins,” Current Opinion in Struc- tural Biology, vol. 16, no. 4, pp. 489–495, 2006.

[145] A. Engel and H. E. Gaub, “Structure and mechanics of membrane proteins,” Annual Review of Biochemistry, vol. 77, pp. 127–148, 2008.

[146] H. J. Butt, K. H. Downing, and P. K. Hansma, “Imaging the membrane-protein bacteriorhodopsin with the atomic force microscope,” Biophysical Journal, vol. 58, no. 6, pp. 1473–1480, 1990.

[147] S. Scheuring, J. Seguin, S. Marco, D. Levy, B. Robert, and J. L. Rigaud, “Nanodis- section and high-resolution imaging of the Rhodopseudomonas viridis photosynthetic core complex in native membranes by AFM,” Proceedings of the National Academy of Sciences of the United States of America, vol. 100, no. 4, pp. 1690–1693, 2003.

[148] H. G. Hansma, J. Vesenka, C. Siegerist, G. Kelderman, H. Morrett, R. L. Sinsheimer, V. Elings, C. Bustamante, and P. K. Hansma, “Reproducible imaging and dissection of plasmid DNA under liquid with the atomic force microscope,” Science, vol. 256, no. 5060, pp. 1180–1184, 1992.

[149] H. G. Hansma, K. J. Kim, D. E. Laney, R. A. Garcia, M. Argaman, M. J. Allen, and S. M. Parsons, “Properties of biomolecules measured from atomic force microscope images: A review,” Journal of Structural Biology, vol. 119, no. 2, pp. 99–108, 1997.

[150] Y. L. Lyubchenko and L. S. Shlyakhtenko, “Visualization of supercoiled DNA with atomic force microscopy in situ,” Proceedings of the National Academy of Sciences of the United States of America, vol. 94, no. 2, pp. 496–501, 1997.

[151] J. Moukhtar, E. Fontaine, C. Faivre-Moskalenko, and A. Arneodo, “Probing persistence in DNA curvature properties with atomic force microscopy,” Physical Review Letters, vol. 98, no. 17, pp. 1–4, 2007.

[152] Y. F. Dufrene and G. U. Lee, “Advances in the characterization of supported lipid ﬁlms with the atomic force microscope,” Biochimica Et Biophysica Acta-Biomembranes, vol. 1509, no. 1-2, pp. 14–41, 2000.

[153] J. Domke, S. Dannohl, W. J. Parak, O. Muller, W. K. Aicher, and M. Radmacher, “Substrate dependent diﬀerences in morphology and elasticity of living osteoblasts investigated by atomic force microscopy,” Colloids and Surfaces B-Biointerfaces, vol. 19, no. 4, pp. 367–379, 2000.

 BIBLIOGRAPHY

[154] R. J. Kroeze, M. N. Helder, W. H. Roos, G. J. Wuite, R. A. Bank, and T. H. Smit, “The eﬀect of ethylene oxide, glow discharge and electron beam on the surface characteristics of poly(l-lactide-co-caprolactone) and the corresponding cellular response of adipose stem cells,” Acta Biomater, vol. 6, no. 6, pp. 2060–5, 2010.

[155] M. Bonomini, B. Pavone, V. Sirolli, F. Del Buono, M. Di Cesare, P. Del Boccio, L. Amoroso, C. Di Ilio, P. Sacchetta, G. Federici, and A. Urbani, “Proteomics characterization of protein adsorption onto hemodialysis membranes,” Journal of Proteome Research, vol. 5, no. 10, pp. 2666–2674, 2006.

[156] N. H. Thomson, M. Fritz, M. Radmacher, J. P. Cleveland, C. F. Schmidt, and P. K. Hansma, “Protein tracking and detection of protein motion using atomic force microscopy,” Biophysical Journal, vol. 70, no. 5, pp. 2421–2431, 1996.

[157] D. J. Muller, A. Engel, U. Matthey, T. Meier, P. Dimroth, and K. Suda, “Observing membrane protein diﬀusion at subnanometer resolution,” Journal of Molecular Biology, vol. 327, no. 5, pp. 925–930, 2003.

[158] M. B. Viani, L. I. Pietrasanta, J. B. Thompson, A. Chand, I. C. Gebeshuber, J. H. Kindt, M. Richter, H. G. Hansma, and P. K. Hansma, “Probing protein-protein interactions in real time,” Nature Structural Biology, vol. 7, no. 8, pp. 644–647, 2000.

[159] M. Yokokawa, C. Wada, T. Ando, N. Sakai, A. Yagi, S. H. Yoshimura, and K. Takeyasu, “Fast-scanning atomic force microscopy reveals the ATP/ADP-dependent conformational changes of GroEL,” EMBO Journal, vol. 25, no. 19, pp. 4567–4576, 2006.

[160] N. H. Thomson, S. Kasas, B. M. Riederer, S. Catsicas, G. Dietler, A. J. Kulik, and L. Forro, “Large ﬂuctuations in the disassembly rate of microtubules revealed by atomic force microscopy,” Ultramicroscopy, vol. 97, no. 1-4, pp. 239–247, 2003.

[161] S. Kasas, N. H. Thomson, B. L. Smith, H. G. Hansma, X. S. Zhu, M. Guthold, C. Bus- tamante, E. T. Kool, M. Kashlev, and P. K. Hansma, “Escherichia coli RNA polymerase activity observed using atomic force microscopy,” Biochemistry, vol. 36, no. 3, pp. 461– 468, 1997.

[162] M. Guthold, X. S. Zhu, C. Rivetti, G. L. Yang, N. H. Thomson, S. Kasas, H. G. Hansma, B. Smith, P. K. Hansma, and C. Bustamante, “Direct observation of one- dimensional diﬀusion and transcription by escherichia coli RNA polymerase,” Biophys- ical Journal, vol. 77, no. 4, pp. 2284–2294, 1999.

[163] E. M. Puchner and H. E. Gaub, “Force and function: probing proteins with AFM-based force spectroscopy,” Current Opinion in Structural Biology, vol. 19, no. 5, pp. 605–614, 2009.

 BIBLIOGRAPHY

[164] M. Radmacher, M. Fritz, C. M. Kacher, J. P. Cleveland, and P. K. Hansma, “Measur- ing the viscoelastic properties of human platelets with the atomic force microscope,” Biophysical Journal, vol. 70, no. 1, pp. 556–567, 1996.

[165] C. Rotsch, K. Jacobson, and M. Radmacher, “Dimensional and mechanical dynamics of active and stable edges in motile ﬁbroblasts investigated by using atomic force microscopy,” Proceedings of the National Academy of Sciences of the United States of America, vol. 96, no. 3, pp. 921–926, 1999.

[166] R. E. Mahaﬀy, C. K. Shih, F. C. MacKintosh, and J. Kas, “Scanning probe-based frequency-dependent microrheology of polymer gels and biological cells,” Physical Re- view Letters, vol. 85, no. 4, pp. 880–883, 2000.

[167] A. Vinckier, C. Dumortier, Y. Engelborghs, and L. Hellemans, “Dynamical and mechanical study of immobilized microtubules with atomic force microscopy,” Journal of Vacuum Science & Technology B, vol. 14, no. 2, pp. 1427–1431, 1996.

[168] A. Kis, S. Kasas, B. Babic, A. J. Kulik, W. Benoit, G. A. D. Briggs, C. Schonenberger, S. Catsicas, and L. Forro, “Nanomechanics of microtubules,” Physical Review Letters, vol. 89, no. 24, pp. 1–4, 2002.

[169] I. L. Ivanovska, P. J. de Pablo, B. Ibarra, G. Sgalari, F. C. MacKintosh, J. L. Car- rascosa, C. F. Schmidt, and G. J. L. Wuite, “Bacteriophage capsids: Tough nanoshells with complex elastic properties,” Proceedings of the National Academy of Sciences of the United States of America, vol. 101, no. 20, pp. 7600–7605, 2004.

[170] M. Radmacher, M. Fritz, J. P. Cleveland, D. A. Walters, and P. K. Hansma, “Imaging adhesion forces and elasticity of lysozyme adsorbed on mica with the atomic-force microscope,” Langmuir, vol. 10, no. 10, pp. 3809–3814, 1994.

[171] A. Ikai, R. Afrin, and H. Sekiguchi, “Pulling and pushing protein molecules by AFM,” Current Nanoscience, vol. 3, no. 1, pp. 17–29, 2007.

[172] E. L. Florin, V. T. Moy, and H. E. Gaub, “Adhesion forces between individual ligand- receptor pairs,” Science, vol. 264, no. 5157, pp. 415–417, 1994.

[173] G. U. Lee, D. A. Kidwell, and R. J. Colton, “Sensing discrete streptavidin biotin interactions with atomic-force microscopy,” Langmuir, vol. 10, no. 2, pp. 354–357, 1994.

[174] R. Merkel, P. Nassoy, A. Leung, K. Ritchie, and E. Evans, “Energy landscapes of receptor-ligand bonds explored with dynamic force spectroscopy,” Nature, vol. 397, no. 6714, pp. 50–53, 1999.

[175] M. Rief, M. Gautel, F. Oesterhelt, J. M. Fernandez, and H. E. Gaub, “Reversible unfolding of individual titin immunoglobulin domains by AFM,” Science, vol. 276, no. 5315, pp. 1109–1112, 1997.

 BIBLIOGRAPHY

[176] G. U. Lee, L. A. Chrisey, and R. J. Colton, “Direct measurement of the forces between complementary strands of DNA,” Science, vol. 266, no. 5186, pp. 771–773, 1994.

[177] D. Fotiadis, S. Scheuring, S. A. Muller, A. Engel, and D. J. Muller, “Imaging and manipulation of biological structures with the AFM,” Micron, vol. 33, no. 4, pp. 385– 397, 2002.

[178] R. J. Colton, “Nanoscale measurements and manipulation,” Journal of Vacuum Science & Technology B, vol. 22, no. 4, pp. 1609–1635, 2004.

[179] K. D. Costa and F. C. P. Yin, “Analysis of indentation: Implications for measuring mechanical properties with atomic force microscopy,” Journal of Biomechanical Engineering-Transactions of the Asme, vol. 121, no. 5, pp. 462–471, 1999.

[180] W. H. Roos, I. L. Ivanovska, A. Evilevitch, and G. J. L. Wuite, “Viral capsids: Me- chanical characteristics, genome packaging and delivery mechanisms,” Cellular and Molecular Life Sciences, vol. 64, no. 12, pp. 1484–1497, 2007.

[181] W. M. Gelbart and C. M. Knobler, “Pressurized viruses,” Science, vol. 323, no. 5922, pp. 1682–1683, 2009.

[182] A. Arkhipov, W. H. Roos, G. J. L. Wuite, and K. Schulten, “Elucidating the mechanism behind irreversible deformation of viral capsids,” Biophysical Journal, vol. 97, no. 7, pp. 2061–2069, 2009.

[183] J. P. Michel, I. L. Ivanovska, M. M. Gibbons, W. S. Klug, C. M. Knobler, G. J. L. Wuite, and C. F. Schmidt, “Nanoindentation studies of full and empty viral capsids and the eﬀects of capsid protein mutations on elasticity and strength,” Proceedings of the National Academy of Sciences of the United States of America, vol. 103, no. 16, pp. 6184–6189, 2006.

[184] C. Uetrecht, C. Versluis, N. R. Watts, W. H. Roos, G. J. L. Wuite, P. T. Wingﬁeld, A. C. Steven, and A. J. R. Heck, “High-resolution mass spectrometry of viral assemblies: Molecular composition and stability of dimorphic hepatitis B virus capsids,” Proceedings of the National Academy of Sciences of the United States of America, vol. 105, no. 27, pp. 9216–9220, 2008.

[185] C. Carrasco, A. Carreira, I. A. T. Schaap, P. A. Serena, J. Gomez-Herrero, M. G. Ma- teu, and P. J. Pablo, “DNA-mediated anisotropic mechanical reinforcement of a virus,” Proceedings of the National Academy of Sciences of the United States of America, vol. 103, no. 37, pp. 13706–13711, 2006.

[186] I. Ivanovska, G. Wuite, B. Jonsson, and A. Evilevitch, “Internal DNA pressure modiﬁes stability of WT phage,” Proceedings of the National Academy of Sciences of the United States of America, vol. 104, no. 23, pp. 9603–9608, 2007.

 BIBLIOGRAPHY

[187] I. Liashkovich, W. Hafezi, J. E. Kuhn, H. Oberleithner, A. Kramer, and V. Shahin, “Exceptional mechanical and structural stability of HSV-1 unveiled with ﬂuid atomic force microscopy,” Journal of Cell Science, vol. 121, no. 14, pp. 2287–2292, 2008.

[188] M. M. Gibbons and W. S. Klug, “Mechanical modeling of viral capsids,” Journal of Materials Science, vol. 42, no. 21, pp. 8995–9004, 2007.

[189] M. Buenemann and P. Lenz, “Mechanical limits of viral capsids,” Proceedings of the Na- tional Academy of Sciences of the United States of America, vol. 104, no. 24, pp. 9925– 9930, 2007.

[190] M. M. Gibbons and W. S. Klug, “Inﬂuence of nonuniform geometry on nanoindentation of viral capsids,” Biophysical Journal, vol. 95, no. 8, pp. 3640–3649, 2008.

[191] N. Kol, M. Gladnikoﬀ, D. Barlam, R. Z. Shneck, A. Rein, and I. Rousso, “Mechani- cal properties of murine leukemia virus particles: Eﬀect of maturation,” Biophysical Journal, vol. 91, no. 2, pp. 767–774, 2006.

[192] Y. Zhao, Z. B. Ge, and J. Y. Fang, “Elastic modulus of viral nanotubes,” Physical Review E, vol. 78, no. 3, p. 031914, 2008.

[193] K. M. Munson, P. G. Mulugeta, and Z. J. Donhauser, “Enhanced mechanical stability of microtubules polymerized with a slowly hydrolyzable nucleotide analogue,” Journal of Physical Chemistry B, vol. 111, no. 19, pp. 5053–5057, 2007.

[194] I. A. T. Schaap, B. Hoffmann, C. Carrasco, R. Merkel, and C. F. Schmidt, “Tau protein binding forms a 1 nm thick layer along protofilaments without affecting the radial elasticity of microtubules,” Journal of Structural Biology, vol. 158, no. 3, pp. 282–292, 2007.

[195] S. Kasas, A. Kis, B. M. Riederer, L. Forro, G. Dietler, and S. Catsicas, “Mechanical properties of microtubules explored using the ﬁnite elements method,” Chemphyschem, vol. 5, no. 2, pp. 252–257, 2004.

[196] C. Guzman, S. Jeney, L. Kreplak, S. Kasas, A. J. Kulik, U. Aebi, and L. Forro, “Ex- ploring the mechanical properties of single vimentin intermediate ﬁlaments by atomic force microscopy,” Journal of Molecular Biology, vol. 360, no. 3, pp. 623–630, 2006.

[197] L. Yang, K. O. van der Werf, B. Koopman, V. Subramaniam, M. L. Bennink, P. J. Dijkstra, and J. Feijen, “Micromechanical bending of single collagen ﬁbrils using atomic force microscopy,” Journal of Biomedical Materials Research Part A, vol. 82A, no. 1, pp. 160–168, 2007.

[198] L. Yang, K. O. Van der Werf, C. F. C. Fitie, M. L. Bennink, P. J. Dijkstra, and J. Feijen, “Mechanical properties of native and cross-linked type i collagen ﬁbrils,” Biophysical Journal, vol. 94, no. 6, pp. 2204–2211, 2008.

 BIBLIOGRAPHY

[199] M. Koenders, L. Yang, R. G. Wismans, K. O. van der Werf, D. P. Reinhardt, W. Daa- men, M. L. Bennink, P. J. Dijkstra, T. H. van Kuppevelt, and J. Feijen, “Microscale mechanical properties of single elastic fibers: The role of fibrillin-microfibrils,” Bioma- terials, vol. 30, no. 13, pp. 2425–2432, 2009.

[200] I. A. T. Schaap, C. Carrasco, P. J. de Pablo, F. C. MacKintosh, and C. F. Schmidt, “Elastic response, buckling, and instability of microtubules under radial indentation,” Biophysical Journal, vol. 91, no. 4, pp. 1521–1531, 2006.

[201] D. Sept and F. C. MacKintosh, “Microtubule elasticity: Connecting all-atom simulations with continuum mechanics,” Physical Review Letters, vol. 104, no. 1, p. 4, 2010.

[202] R. Afrin, M. T. Alam, and A. Ikai, “Pretransition and progressive softening of bovine carbonic anhydrase ii as probed by single molecule atomic force microscopy,” Protein Science, vol. 14, no. 6, pp. 1447–1457, 2005.

[203] T. Kodama, H. Ohtani, H. Arakawa, and A. Ikai, “Mechanical perturbation-induced ﬂuorescence change of green ﬂuorescent protein,” Applied Physics Letters, vol. 86, no. 4, p. 3, 2005.

[204] T. Kodama, H. Ohtani, H. Arakawa, and A. Ikai, “Observation of the destruction of biomolecules under compression force,” Ultramicroscopy, vol. 105, no. 1-4, pp. 189–195, 2005.

[205] M. E. Browning-Kelley, K. Wadu-Mesthrige, V. Hari, and G. Y. Liu, “Atomic force microscopic study of speciﬁc antigen/antibody binding,” Langmuir, vol. 13, no. 2, pp. 343–350, 1997.

[206] S. Lin, J.-L. Chen, L.-S. Huang, and H.-W. Lin, “Measurements of the forces in protein interactions with atomic force microscopy,” Current Proteomics, vol. 2, pp. 55–81, 2005.

[207] V. T. Moy, E. L. Florin, and H. E. Gaub, “Intermolecular forces and energies between ligands and receptors,” Science, vol. 266, no. 5183, pp. 257–259, 1994.

[208] H. Grubmuller, B. Heymann, and P. Tavan, “Ligand binding: Molecular mechanics calculation of the streptavidin biotin rupture force,” Science, vol. 271, no. 5251, pp. 997– 999, 1996.

[209] E. Evans and K. Ritchie, “Dynamic strength of molecular adhesion bonds,” Biophysical Journal, vol. 72, no. 4, pp. 1541–1555, 1997.

[210] S. Izrailev, S. Stepaniants, M. Balsera, Y. Oono, and K. Schulten, “Molecular dynamics study of unbinding of the avidin-biotin complex,” Biophysical Journal, vol. 72, no. 4, pp. 1568–1581, 1997.

 BIBLIOGRAPHY

[211] O. H. Willemsen, M. M. E. Snel, K. O. van der Werf, B. G. de Grooth, J. Greve, P. Hinterdorfer, H. J. Gruber, H. Schindler, Y. van Kooyk, and C. G. Figdor, “Simul- taneous height and adhesion imaging of antibody-antigen interactions by atomic force microscopy,” Biophysical Journal, vol. 75, no. 5, pp. 2220–2228, 1998.

[212] H. Nakajima, Y. Kunioka, K. Nakano, K. Shimizu, M. Seto, and T. Ando, “Scan- ning force microscopy of the interaction events between a single molecule of heavy meromyosin and actin,” Biochemical and Biophysical Research Communications, vol. 234, no. 1, pp. 178–182, 1997.

[213] E. M. Puchner, A. Alexandrovich, A. L. Kho, U. Hensen, L. V. Schafer, B. Brand- meier, F. Grater, H. Grubmuller, H. E. Gaub, and M. Gautel, “Mechanoenzymatics of titin kinase,” Proceedings of the National Academy of Sciences of the United States of America, vol. 105, no. 36, pp. 13385–13390, 2008.

[214] R. Perez-Jimenez, S. Garcia-Manyes, S. R. K. Ainavarapu, and J. M. Fernandez, “Me- chanical unfolding pathways of the enhanced yellow ﬂuorescent protein revealed by single molecule force spectroscopy,” Journal of Biological Chemistry, vol. 281, no. 52, pp. 40010–40014, 2006.

[215] H. Dietz and M. Rief, “Exploring the energy landscape of GFP by single-molecule mechanical experiments,” Proceedings of the National Academy of Sciences of the United States of America, vol. 101, no. 46, pp. 16192–16197, 2004.

[216] T. Bornschlogl, G. Woehlke, and M. Rief, “Single molecule mechanics of the kinesin neck,” Proceedings of the National Academy of Sciences of the United States of Amer- ica, vol. 106, no. 17, pp. 6992–6997, 2009.

[217] H. Lu, B. Isralewitz, A. Krammer, V. Vogel, and K. Schulten, “Unfolding of titin immunoglobulin domains by steered molecular dynamics simulation,” Biophysical Jour- nal, vol. 75, no. 2, pp. 662–671, 1998.

[218] P. E. Marszalek, H. Lu, H. B. Li, M. Carrion-Vazquez, A. F. Oberhauser, K. Schulten, and J. M. Fernandez, “Mechanical unfolding intermediates in titin modules,” Nature, vol. 402, no. 6757, pp. 100–103, 1999.

[219] F. Grater, J. H. Shen, H. L. Jiang, M. Gautel, and H. Grubmuller, “Mechanically induced titin kinase activation studied by force-probe molecular dynamics simulations,” Biophysical Journal, vol. 88, no. 2, pp. 790–804, 2005.

[220] Y. Hisatomi, D. Katagiri, S. Neya, M. Hara, and T. Hoshino, “Analysis of the unfolding process of green ﬂuorescent protein by molecular dynamics simulation,” Journal of Physical Chemistry B, vol. 112, no. 29, pp. 8672–8680, 2008.

 BIBLIOGRAPHY

[221] D. A. Heuveling, G. W. M. Visser, M. Baclayon, W. H. Roos, G. J. L. Wuite, O. S. Hoekstra, C. R. Leemans, R. de Bree, and G. A. M. S. van Dongen, “89Zr Nanocolloidal Albumin-Based PET/CT Lymphoscintigraphy for Sentinel Node Detection in Head and Neck Cancer: Preclinical Results,” Journal of Nuclear Medicine, vol. 52, no. 10, pp. 1580–1584, 2011.

[222] G. Healthcare, Summary of Product Characteristics Nanocoll. Eindhoven, The Nether- lands, 2010.

[223] X. Jiang, M. Wang, D. Y. Graham, and M. K. Estes, “Expression, self-assembly, and antigenicity of the Norwalk virus capsid protein,” Journal of Virology, vol. 66, no. 11, pp. 6527–32, 1992.

[224] R. H. H. van den Heuvel, E. van Duijn, H. Mazon, S. A. Synowsky, K. Lorenzen, C. Versluis, S. J. J. Brouns, D. Langridge, J. van der Oost, J. Hoyes, and A. J. R. Heck, “Improving the performance of a quadrupole time-of-ﬂight instrument for macromolecular mass spectrometry,” Analytical Chemistry, vol. 78, no. 21, pp. 7473–7483, 2006.

[225] I. Horcas, R. Fernandez, J. M. Gomez-Rodriguez, J. Colchero, J. Gomez-Herrero, and A. M. Baro, “WSXM: A software for scanning probe microscopy and a tool for nanotechnology,” Review of Scientiﬁc Instruments, vol. 78, no. 1, 2007.

[226] G. Farge, M. Mehmedovic, M. Baclayon, S. M. van den Wildenberg, W. H. Roos, C. M. Gustafsson, G. J. Wuite, and M. Falkenberg, “Nucleoid packaging blocks replication and transcription of mitochondrial DNA,” manuscript in preparation, 2013.

[227] N. Larsson and D. Clayton, “Molecular genetic aspects of human mitochondrial disor- ders,” Annual Review of Genetics, vol. 29, pp. 151–178, 1995.

[228] A. Trifunovic, A. Wredenberg, M. Falkenberg, J. Spelbrink, A. Rovio, C. Bruder, M. Bohlooly-Y, S. Gidlof, A. Oldfors, R. Wibom, J. Tornell, H. Jacobs, and N. Lars- son, “Premature ageing in mice expressing defective mitochondrial DNA polymerase,” Nature, vol. 429, no. 6990, pp. 417–423, 2004.

[229] R. P. Fisher and D. A. Clayton, “A transcriptional factor required for promoter recognition by human mitochondrial RNA polymerase - Accurate initiation at the heavy strand and light strand promoters dissected and reconstituted in vitro,” Journal of Biological Chemistry, vol. 260, no. 20, pp. 1330–1338, 1985.

[230] M. Falkenberg, M. Gaspari, A. Rantanen, A. Trifunovic, N. Larsson, and C. Gustafs- son, “Mitochondrial transcription factors B1 and B2 activate transcription of human mtDNA,” Nature Genetics, vol. 31, no. 3, pp. 289–294, 2002.

 BIBLIOGRAPHY

[231] D. Litonin, M. Sologub, Y. Shi, M. Savkina, M. Anikin, M. Falkenberg, C. M. Gustafs- son, and D. Temiakov, “Human Mitochondrial Transcription Revisited: Only TFAM and TFB2M are required for transcription of the mitochondrial genes in vitro,” Journal of Biological Chemistry, vol. 285, no. 24, pp. 18129–18133, 2010.

[232] Y. Shi, A. Dierckx, P. H. Wanrooij, S. Wanrooij, N.-G. Larsson, L. M. Wilhelmsson, M. Falkenberg, and C. M. Gustafsson, “Mammalian transcription factor A is a core component of the mitochondrial transcription machinery,” Proceedings of the National Academy of Sciences of the United States of America, vol. 109, no. 41, pp. 16510– 16515, 2012.

[233] A. Rubio-Cosials, J. F. Sidow, N. Jimenez-Menendez, P. Fernandez-Millan, J. Montoya, H. T. Jacobs, M. Coll, P. Bernado, and M. Sola, “Human mitochondrial transcription factor A induces a U-turn structure in the light strand promoter,” Nature Structural & Molecular Biology, vol. 18, no. 11, pp. 1281–U133, 2011.

[234] C. Takamatsu, S. Umeda, T. Ohsato, T. Ohno, Y. Abe, A. Fukuoh, H. Shinagawa, N. Hamasaki, and D. Kang, “Regulation of mitochondrial D-loops by transcription factor A and single-stranded DNA-binding protein,” EMBO Reports, vol. 3, no. 5, pp. 451–456, 2002.

[235] G. Farge, N. Laurens, O. D. Broekmans, S. M. J. L. van den Wildenberg, L. C. M. Dekker, M. Gaspari, C. M. Gustafsson, E. J. G. Peterman, M. Falkenberg, and G. J. L. Wuite, “Protein sliding and DNA denaturation are essential for DNA organization by human mitochondrial transcription factor A,” Nature Communications, vol. 3, 2012.

[236] M. Ekstrand, M. Falkenberg, A. Rantanen, C. Park, M. Gaspari, K. Hultenby, P. Rustin, C. Gustafsson, and N. Larsson, “Mitochondrial transcription factor A regulates mtDNA copy number in mammals,” Human Molecular Genetics, vol. 13, no. 9, pp. 935–944, 2004.

[237] Y. Matsushima, Y.-i. Goto, and L. S. Kaguni, “Mitochondrial Lon protease regulates mitochondrial DNA copy number and transcription by selective degradation of mitochondrial transcription factor A (TFAM),” Proceedings of the National Academy of Sciences of the United States of America, vol. 107, no. 43, pp. 18410–18415, 2010.

[238] B. Lu, J. Lee, X. Nie, M. Li, Y. I. Morozov, S. Venkatesh, D. F. Bogenhagen, D. Temi- akov, and C. K. Suzuki, “Phosphorylation of Human TFAM in Mitochondria Impairs DNA Binding and Promotes Degradation by the AAA(+) Lon Protease,” Molecular Cell, vol. 49, no. 1, pp. 121–132, 2013.

[239] D. A. Heuveling, G. W. M. Visser, M. de Groot, J. F. de Boer, M. Baclayon, W. H. Roos, G. J. L. Wuite, C. R. Leemans, R. de Bree, and G. A. M. S. van Dongen, “Nanocolloidal albumin-IRDye 800CW: a near-infrared ﬂuorescent tracer with optimal

 BIBLIOGRAPHY

retention in the sentinel lymph node,” European Journal of Nuclear Medicine and Molecular Imaging, vol. 39, no. 7, pp. 1161–1168, 2012.

[240] L. W. T. Alkureishi, G. L. Ross, T. Shoaib, D. S. Soutar, A. G. Robertson, R. Thomp- son, K. D. Hunter, J. A. Sorensen, J. Thomsen, A. Krogdahl, J. Alvarez, L. Barbier, J. Santamaria, T. Poli, E. Sesenna, A. F. Kovacs, F. Gruenwald, L. Barzan, S. Sulfaro, and F. Alberti, “Sentinel Node Biopsy in Head and Neck Squamous Cell Cancer: 5- Year Follow-Up of a European Multicenter Trial,” Annals of Surgical Oncology, vol. 17, no. 9, pp. 2459–2464, 2010.

[241] M. A. Broglie, S. R. Haile, and S. J. Stoeckli, “Long-Term Experience in Sentinel Node Biopsy for Early Oral and Oropharyngeal Squamous Cell Carcinoma,” Annals of Surgical Oncology, vol. 18, no. 10, pp. 2732–2738, 2011.

[242] M. K. Estes, B. V. V. Prasad, and R. L. Atmar, “Noroviruses everywhere: has some- thing changed?,” Current Opinion in Infectious Diseases, vol. 19, no. 5, pp. 467–474, 2006.

[243] K. Green, T. Ando, M. Balayan, T. Berke, I. Clarke, M. Estes, D. Matson, S. Nakata, J. Neill, M. Studdert, and H. Thiel, “Taxonomy of the caliciviruses,” Journal of In- fectious Diseases, vol. 181, no. 2, pp. S322–S330, 2000. International Workshop on Human Caliciviruses, ATLANTA, GEORGIA, MAR 29-31, 1999.

[244] X. Jiang, M. Wang, K. N. Wang, and M. K. Estes, “Sequence and genomic organization of Norwalk virus,” Virology, vol. 195, no. 1, pp. 51–61, 1993.

[245] A. Bertolotti-Ciarlet, L. J. White, R. Chen, B. V. V. Prasad, and M. K. Estes, “Struc- tural requirements for the assembly of Norwalk virus-like particles,” Journal of Virol- ogy, vol. 76, no. 8, pp. 4044–4055, 2002.

[246] A. Bertolotti-Ciarlet, S. E. Crawford, A. M. Hutson, and M. K. Estes, “The 3’ end of Norwalk virus mRNA contains determinants that regulate the expression and stability of the viral capsid protein VP1: a novel function for the VP2 protein,” Journal of Virology, vol. 77, no. 21, pp. 11603–11615, 2003.

[247] L. White, M. Hardy, and H. Estes, “Biochemical characterization of a smaller form of recombinant Norwalk virus capsids assembled in insect cells,” Journal of Virology, vol. 71, no. 10, pp. 8066–8072, 1997.

[248] S. Harrison, “The familiar and the unexpected in structures of icosahedral viruses,” Current Opinion in Structural Biology, vol. 11, no. 2, pp. 195–199, 2001.

[249] D. C. Rapaport, “Role of Reversibility in Viral Capsid Growth: A Paradigm for Self- Assembly,” Physical Review Letters, vol. 101, no. 18, 2008.

 BIBLIOGRAPHY

[250] F. Zenhausern, M. Adrian, R. Emch, M. Taborelli, M. Jobin, and P. Descouts, “Scan- ning force microscopy and cryoelectron microscopy of tobacco mosaic-virus as a test specimen,” Ultramicroscopy, vol. 42, pp. 1168–1172, 1992.

[251] E. Droz, M. Taborelli, T. N. C. Wells, and P. Descouts, “Preparation of isolated biomolecules for SFM observations - bacteriophage-T4 as a test sample,” Biophysi- cal Journal, vol. 65, no. 3, pp. 1180–1187, 1993.

[252] K. Imai, K. Yoshimura, M. Tomitori, O. Nishikawa, R. Kokawa, M. Yamamoto, M. Kobayashi, and A. Ikai, “Scanning tunneling and atomic-force microscopy of T4- bacteriophage and tobacco mosaic-virus,” Japanese Journal of Applied Physics, vol. 32, no. 6B, pp. 2962–2964, 1993.

[253] G. R. Bushell, G. S. Watson, S. A. Holt, and S. Myhra, “Imaging and nano-dissection of tobacco mosaic-virus by atomic-force microscopy,” Journal of Microscopy-Oxford, vol. 180, pp. 174–181, 1995.

[254] M. R. Falvo, S. Washburn, R. Superﬁne, M. Finch, F. P. Brooks, V. Chi, and R. M. Taylor, “Manipulation of individual viruses: Friction and mechanical properties,” Bio- physical Journal, vol. 72, no. 3, pp. 1396–1403, 1997.

[255] A. L. LlamasSaiz, M. AgbandjeMcKenna, W. R. Wikoﬀ, J. Bratton, P. Tattersall, and M. G. Rossmann, “Structure determination of Minute Virus of Mice,” Acta Crystallo- graphica Section D-Biological Crystallography, vol. 53, pp. 93–102, 1997.

[256] J. A. Speir, S. Munshi, G. J. Wang, T. S. Baker, and J. E. Johnson, “Structures of the native and swollen forms of cowpea chlorotic mottle virus determined by x-ray crystallography and cryoelectron microscopy,” Structure, vol. 3, no. 1, pp. 63–78, 1995.

[257] S. R. Casjens and R. W. Hendrix, “Locations and amounts of major structural proteins in bacteriophage-lambda,” Journal of Molecular Biology, vol. 88, no. 2, pp. 535–538, 1974.

[258] T. Dokland and H. Murialdo, “Structural transitions during maturation of bacteriophage lambda capsids,” Journal of Molecular Biology, vol. 233, no. 4, pp. 682–94, 1993.

[259] Y. Z. Tao, N. H. Olson, W. Xu, D. L. Anderson, M. G. Rossmann, and T. S. Baker, “As- sembly of a tailed bacterial virus and its genome release studied in three dimensions,” Cell, vol. 95, no. 3, pp. 431–437, 1998.

[260] F. H. C. Crick and J. D. Watson, “Structure of small viruses,” Nature, vol. 177, no. 4506, pp. 473–475, 1956.

[261] D. L. D. Caspar and A. Klug, “Physical principles in construction of regular viruses,” Cold Spring Harbor Symposia on Quantitative Biology, vol. 27, pp. 1–24, 1962.

 BIBLIOGRAPHY

[262] T. T. Nguyen, R. F. Bruinsma, and W. M. Gelbart, “Elasticity theory and shape transitions of viral shells,” Physical Review E, vol. 72, no. 5, p. 051923, 2005.

[263] K. Namba and G. Stubbs, “Structure of tobacco mosaic-virus at 3.6-a resolution - implications for assembly,” Science, vol. 231, no. 4744, pp. 1401–1406, 1986.

[264] D. L. Anderson, D. D. Hickman, and B. E. Reilly, “Structure of bacillus subtilis bacteriophage phi29 and length of phi29 deoxyribonucleic acid,” Journal of Bacteriology, vol. 91, no. 5, pp. 2081–2089, 1966.

[265] T. Douglas and M. Young, “Host-guest encapsulation of materials by assembled virus protein cages,” Nature, vol. 393, no. 6681, pp. 152–155, 1998.

[266] S. W. Lee, C. B. Mao, C. E. Flynn, and A. M. Belcher, “Ordering of quantum dots using genetically engineered viruses,” Science, vol. 296, no. 5569, pp. 892–895, 2002.

[267] P. Singh, M. J. Gonzalez, and M. Manchester, “Viruses and their uses in nanotechnology,” Drug Development Research, vol. 67, no. 1, pp. 23–41, 2006.

[268] M. Fischlechner and E. Donath, “Viruses as building blocks for materials and devices,” Angewandte Chemie-International Edition, vol. 46, no. 18, pp. 3184–3193, 2007.

[269] P. K. Hansma, J. P. Cleveland, M. Radmacher, D. A. Walters, P. E. Hillner, M. Bezanilla, M. Fritz, D. Vie, H. G. Hansma, C. B. Prater, J. Massie, L. Fuku- naga, J. Gurley, and V. Elings, “Tapping mode atomic-force microscopy in liquids,” Applied Physics Letters, vol. 64, no. 13, pp. 1738–1740, 1994.

[270] C. A. J. Putman, K. O. Vanderwerf, B. G. Degrooth, N. F. Vanhulst, and J. Greve, “Tapping mode atomic-force microscopy in liquid,” Applied Physics Letters, vol. 64, no. 18, pp. 2454–2456, 1994.

[271] X. Xu, C. Carrasco, P. J. de Pablo, J. Gomez-Herrero, and A. Raman, “Unmask- ing imaging forces on soft biological samples in liquids when using dynamic atomic force microscopy: A case study on viral capsids,” Biophysical Journal, vol. 95, no. 5, pp. 2520–2528, 2008.

[272] W. H. Han, S. M. Lindsay, and T. W. Jing, “A magnetically driven oscillating probe microscope for operation in liquids,” Applied Physics Letters, vol. 69, no. 26, pp. 4111– 4113, 1996.

[273] W. H. Han, S. M. Lindsay, M. Dlakic, and R. E. Harington, “Kinked DNA,” Nature, vol. 386, no. 6625, pp. 563–563, 1997.

[274] A. Engel, “Biological applications of scanning probe microscopes,” Annual Review of Biophysics and Biophysical Chemistry, vol. 20, pp. 79–108, 1991.

 BIBLIOGRAPHY

[275] V. J. Morris, “Biological applications of scanning probe microscopies,” Progress in Biophysics & Molecular Biology, vol. 61, no. 2, pp. 131–185, 1994.

[276] C. Bustamante, C. Rivetti, and D. J. Keller, “Scanning force microscopy under aqueous solutions,” Current Opinion in Structural Biology, vol. 7, no. 5, pp. 709–716, 1997.

[277] W. R. Bowen, R. W. Lovitt, and C. J. Wright, “Application of atomic force microscopy to the study of micromechanical properties of biological materials,” Biotechnology Let- ters, vol. 22, no. 11, pp. 893–903, 2000.

[278] M. Gaczynska and P. A. Osmulski, “AFM of biological complexes: What can we learn?,” Current Opinion in Colloid & Interface Science, vol. 13, no. 5, pp. 351–367, 2008.

[279] J. Vesenka, S. Manne, R. Giberson, T. Marsh, and E. Henderson, “Colloidal gold particles as an incompressible atomic-force microscope imaging standard for assessing the compressibility of biomolecules,” Biophysical Journal, vol. 65, no. 3, pp. 992–997, 1993.

[280] A. Ikai, K. Yoshimura, F. Arisaka, A. Ritani, and K. Imai, “Atomic-force microscopy of bacteriophage-T4 and its tube-baseplate complex,” FEBS Letters, vol. 326, no. 1-3, pp. 39–41, 1993.

[281] Y. L. Lyubchenko, P. I. Oden, D. Lampner, S. M. Lindsay, and K. A. Dunker, “Atomic force microscopy of DNA and bacteriophage in air, water and propanol - the role of adhesion forces,” Nucleic Acids Research, vol. 21, no. 5, pp. 1117–1123, 1993.

[282] S. Karrasch, M. Dolder, F. Schabert, J. Ramsden, and A. Engel, “Covalent binding of biological samples to solid supports for scanning probe microscopy in buﬀer solution,” Biophysical Journal, vol. 65, no. 6, pp. 2437–2446, 1993.

[283] D. Anselmetti, R. Luthi, E. Meyer, T. Richmond, M. Dreier, J. E. Frommer, and H. J. Guntherodt, “Attractive-mode imaging of biological-materials with dynamic force microscopy,” Nanotechnology, vol. 5, no. 2, pp. 87–94, 1994.

[284] D. Anselmetti, M. Dreier, R. Luthi, T. Richmond, E. Meyer, J. Frommer, and H. J. Guntherodt, “Biological-materials studied with dynamic force microscopy,” Journal of Vacuum Science & Technology B, vol. 12, pp. 1500–1503, 1994.

[285] A. J. Malkin, Y. G. Kuznetsov, T. A. Land, J. J. Deyoreo, and A. McPherson, “Mech- anisms of growth for protein and virus crystals,” Nature Structural Biology, vol. 2, no. 11, pp. 956–959, 1995.

[286] A. J. Malkin, T. A. Land, Y. G. Kuznetsov, A. McPherson, and J. J. Deyoreo, “In- vestigation of virus crystal-growth mechanisms by in-situ atomic-force microscopy,” Physical Review Letters, vol. 75, no. 14, pp. 2778–2781, 1995.

 BIBLIOGRAPHY

[287] R. W. Lucas, Y. G. Kuznetsov, S. B. Larson, and A. McPherson, “Crystallization of brome mosaic virus and T=1 brome mosaic virus particles following a structural transition,” Virology, vol. 286, no. 2, pp. 290–303, 2001.

[288] A. J. Malkin, Y. G. Kuznetsov, R. W. Lucas, and A. McPherson, “Surface processes in the crystallization of turnip yellow mosaic virus visualized by atomic force microscopy,” Journal of Structural Biology, vol. 127, no. 1, pp. 35–43, 1999.

[289] V. V. Peremyslov, I. A. Andreev, A. I. Prokhnevsky, G. H. Duncan, M. E. Taliansky, and V. V. Dolja, “Complex molecular architecture of beet yellows virus particles,” Proceedings of the National Academy of Sciences of the United States of America, vol. 101, no. 14, pp. 5030–5035, 2004.

[290] B. La Scola, S. Audic, C. Robert, L. Jungang, X. de Lamballerie, M. Drancourt, R. Birtles, J. M. Claverie, and D. Raoult, “A giant virus in amoebae,” Science, vol. 299, no. 5615, pp. 2033–2033, 2003.

[291] C. A. Xiao, P. R. Chipman, A. J. Battisti, V. D. Bowman, P. Renesto, D. Raoult, and M. G. Rossmann, “Cryo-electron microscopy of the giant mimivirus,” Journal of Molecular Biology, vol. 353, no. 3, pp. 493–496, 2005.

[292] C. Xiao, Y. G. Kuznetsov, S. Y. Sun, S. L. Hafenstein, V. A. Kostyuchenko, P. R. Chip- man, M. Suzan-Monti, D. Raoult, A. McPherson, and M. G. Rossmann, “Structural studies of the Giant Mimivirus,” Plos Biology, vol. 7, no. 4, pp. 958–966, 2009.

[293] N. Matsko, D. Klinov, A. Manykin, V. Demin, and S. Klimenko, “Atomic force microscopy analysis of bacteriophages phi KZ and T4,” Journal of Electron Microscopy, vol. 50, no. 5, pp. 417–422, 2001.

[294] A. Fokine, V. A. Kostyuchenko, A. V. Eﬁmov, L. P. Kurochkina, N. N. Sykilinda, J. Robben, G. Volckaert, A. Hoenger, P. R. Chipman, A. J. Battisti, M. G. Rossmann, and V. V. Mesyanzhinov, “A three-dimensional cryo-electron microscopy structure of the bacteriophage phi KZ head,” Journal of Molecular Biology, vol. 352, no. 1, pp. 117– 124, 2005.

[295] P. Boevink and K. J. Oparka, “Virus-host interactions during movement processes,” Plant Physiology, vol. 138, no. 4, pp. 1815–1821, 2005.

[296] W. J. Lucas, “Plant viral movement proteins: Agents for cell-to-cell traﬃcking of viral genomes,” Virology, vol. 344, no. 1, pp. 169–184, 2006.

[297] O. I. Kiselyova, I. V. Yaminsky, O. V. Karpova, N. P. Rodionova, S. V. Kozlovsky, M. V. Arkhipenko, and J. G. Atabekov, “AFM study of potato virus X disassembly induced by movement protein,” Journal of Molecular Biology, vol. 332, no. 2, pp. 321– 325, 2003.

 BIBLIOGRAPHY

[298] C. L. Cheung, S. W. Chung, A. Chatterji, T. W. Lin, J. E. Johnson, S. Hok, J. Perkins, and J. J. De Yoreo, “Physical controls on directed virus assembly at nanoscale chemical templates,” Journal of the American Chemical Society, vol. 128, no. 33, pp. 10801– 10807, 2006.

[299] F. Kienberger, C. Rankl, V. Pastushenko, R. Zhu, D. Blaas, and P. Hinterdorfer, “Vi- sualization of single receptor molecules bound to human rhinovirus under physiological conditions,” Structure, vol. 13, no. 9, pp. 1247–1253, 2005.

[300] M. Plomp, M. K. Rice, E. K. Wagner, A. McPherson, and A. J. Malkin, “Rapid visualization at high resolution of pathogens by atomic force microscopy - structural studies of herpes simplex virus-1,” American Journal of Pathology, vol. 160, no. 6, pp. 1959–1966, 2002.

[301] W. W. Newcomb and J. C. Brown, “Structure of the herpes-simplex virus capsid - effects of extraction with guanidine-hydrochloride and partial reconstitution of extracted capsids,” Journal of Virology, vol. 65, no. 2, pp. 613–620, 1991.

[302] W. W. Newcomb, B. L. Trus, F. P. Booy, A. C. Steven, J. S. Wall, and J. C. Brown, “Structure of the herpes simplex virus capsid. molecular composition of the pentons and the triplexes,” Journal of Molecular Biology, vol. 232, no. 2, pp. 499–511, 1993.

[303] A. C. Steven, C. R. Roberts, J. Hay, M. E. Bisher, T. Pun, and B. L. Trus, “Hex- avalent capsomers of herpes-simplex virus type-2 - symmetry, shape, dimensions, and oligomeric status,” Journal of Virology, vol. 57, no. 2, pp. 578–584, 1986.

[304] J. B. Heymann, N. Q. Cheng, W. W. Newcomb, B. L. Trus, J. C. Brown, and A. C. Steven, “Dynamics of herpes simplex virus capsid maturation visualized by time-lapse cryo-electron microscopy,” Nature Structural Biology, vol. 10, no. 5, pp. 334–341, 2003.

[305] L. Ponchon, S. Mangenot, P. Boulanger, and L. Letellier, “Encapsidation and transfer of phage DNA into host cells: From in vivo to single particles studies,” Biochimica Et Biophysica Acta-General Subjects, vol. 1724, no. 3, pp. 255–261, 2005.

[306] M. Marsh and A. Helenius, “Virus entry: Open sesame,” Cell, vol. 124, no. 4, pp. 729– 740, 2006.

[307] Y. F. Drygin, O. A. Bordunova, M. O. Gallyamov, and I. V. Yaminsky, “Atomic force microscopy examination of tobacco mosaic virus and virion RNA,” FEBS Letters, vol. 425, no. 2, pp. 217–221, 1998.

[308] Y. G. Kuznetsov and A. McPherson, “Atomic force microscopy investigation of Turnip Yellow Mosaic Virus capsid disruption and RNA extrusion,” Virology, vol. 352, no. 2, pp. 329–337, 2006.

 BIBLIOGRAPHY

[309] C. Carrasco, M. Douas, R. Miranda, M. Castellanos, P. A. Serena, J. L. Carrascosa, M. G. Mateu, M. I. Marques, and P. J. de Pablo, “The capillarity of nanometric water menisci conﬁned inside closed-geometry viral cages,” Proceedings of the National Academy of Sciences of the United States of America, vol. 106, no. 14, pp. 5475–5480, 2009.

[310] V. Shahin, W. Hafezi, H. Oberleithner, Y. Ludwig, B. Windoﬀer, H. Schillers, and J. E. Kuhn, “The genome of HSV-1 translocates through the nuclear pore as a condensed rod-like structure,” Journal of Cell Science, vol. 119, no. 1, pp. 23–30, 2006.

[311] E. Prchla, E. Kuechler, D. Blaas, and R. Fuchs, “Uncoating of human rhinovirus serotype-2 from late endosomes,” Journal of Virology, vol. 68, no. 6, pp. 3713–3723, 1994.

[312] E. A. Hewat, E. Neumann, and D. Blaas, “The concerted conformational changes during human rhinovirus 2 uncoating,” Molecular Cell, vol. 10, no. 2, pp. 317–326, 2002.

[313] J. K. H. Horber, W. Haberle, F. Ohnesorge, G. Binnig, H. G. Liebich, C. P. Czerny, H. Mahnel, and A. Mayr, “Investigation of living cells in the nanometer regime with the scanning force microscope,” Scanning Microscopy, vol. 6, no. 4, pp. 919–930, 1992.

[314] G. V. Stokes, “High-voltage electron-microscope study of release of vaccinia virus from whole cells,” Journal of Virology, vol. 18, no. 2, pp. 636–643, 1976.

[315] Y. G. Kuznetsov, S. Datta, N. H. Kothari, A. Greenwood, H. Fan, and A. McPher- son, “Atomic force microscopy investigation of ﬁbroblasts infected with wild-type and mutant murine leukemia virus (MuLV),” Biophysical Journal, vol. 83, no. 6, pp. 3665– 3674, 2002.

[316] A. Low, S. Datta, Y. Kuznetsov, S. Jahid, N. Kothari, A. McPherson, and H. Fan, “Mutation in the glycosylated gag protein of murine leukemia virus results in reduced in vivo infectivity and a novel defect in viral budding or release,” Journal of Virology, vol. 81, no. 8, pp. 3685–3692, 2007.

[317] L. V. Chernomordik and M. M. Kozlov, “Protein-lipid interplay in fusion and ﬁssion of biological membranes,” Annual Review of Biochemistry, vol. 72, pp. 175–207, 2003.

[318] T. M. Svitkina and G. G. Borisy, “Arp2/3 complex and actin depolymerizing factor coﬁlin in dendritic organization and treadmilling of actin ﬁlament array in lamellipo- dia,” Journal of Cell Biology, vol. 145, no. 5, pp. 1009–1026, 1999.

[319] D. A. Fletcher and D. Mullins, “Cell mechanics and the cytoskeleton,” Nature, vol. 463, no. 7280, pp. 485–492, 2010.

[320] Y. G. Kuznetsov, J. G. Victoria, W. E. Robinson, and A. McPherson, “Atomic force microscopy investigation of human immunodeﬁciency virus (HIV) and HIV-infected lymphocytes,” Journal of Virology, vol. 77, no. 22, pp. 11896–11909, 2003.

 BIBLIOGRAPHY

[321] M. Moloney, L. McDonnell, and H. O’Shea, “Atomic force microscopy analysis of enveloped and non-enveloped viral entry into, and egress from, cultured cells,” Ultrami- croscopy, vol. 100, no. 3-4, pp. 163–9, 2004.

[322] F. Kienberger, R. Moser, H. Schindler, D. Blaas, and P. Hinterdorfer, “Quasi-crystalline arrangement of human rhinovirus 2 on model cell membranes,” Single Molecules, vol. 2, pp. 99–103, 2001.

[323] E. V. Dubrovin, A. G. Voloshin, S. V. Kraevsky, T. E. Ignatyuk, S. S. Abramchuk, I. V. Yaminsky, and S. G. Ignatov, “Atomic force microscopy investigation of phage infection of bacteria,” Langmuir, vol. 24, no. 22, pp. 13068–13074, 2008.

[324] Y. Y. Chen, C. C. Wu, J. L. Hsu, H. L. Peng, H. Y. Chang, and T. R. Yew, “Surface rigidity change of escherichia coli after ﬁlamentous bacteriophage infection,” Langmuir, vol. 25, no. 8, pp. 4607–4614, 2009.

[325] N. A. Amro, L. P. Kotra, K. Wadu-Mesthrige, A. Bulychev, S. Mobashery, and G. Y. Liu, “High-resolution atomic force microscopy studies of the escherichia coli outer membrane: Structural basis for permeability,” Langmuir, vol. 16, no. 6, pp. 2789–2796, 2000.

[326] A. Roy and S. Mitra, “Increased fragility of escherichia-coli after infection with bacte- riphage M13,” Journal of Virology, vol. 6, no. 3, pp. 333–339, 1970.

[327] S. Mitragotri and J. Lahann, “Physical approaches to biomaterial design,” Nature Ma- terials, vol. 8, no. 1, pp. 15–23, 2009.

[328] P. E. Sheehan and C. M. Lieber, “Nanotribology and nanofabrication of MoO3 structures by atomic force microscopy,” Science, vol. 272, no. 5265, pp. 1158–1161, 1996.

[329] E. Meyer, R. Overney, D. Brodbeck, L. Howald, R. Luthi, J. Frommer, and H. J. Guntherodt, “Friction and wear of langmuir-blodgett-ﬁlms observed by friction force microscopy,” Physical Review Letters, vol. 69, no. 12, pp. 1777–1780, 1992.

[330] J. Y. Fang, C. M. Knobler, M. Gingery, and F. A. Eiserling, “Imaging bacteriophage t4 on patterned organosilane monolayers by scanning force microscopy,” Journal of Physical Chemistry B, vol. 101, no. 43, pp. 8692–8695, 1997.

[331] M. Guthold, M. R. Falvo, W. G. Matthews, S. Paulson, S. Washburn, D. A. Erie, R. Superﬁne, F. P. Brooks, and R. M. Taylor, “Controlled manipulation of molecular samples with the nanomanipulator,” in IEEE-ASME Transactions on Mechatronics, vol. 5, pp. 189–198, Ieee-Inst Electrical Electronics Engineers Inc, 2000.

[332] A. Schmatulla, N. Maghelli, and O. Marti, “Micromechanical properties of tobacco mosaic viruses,” Journal of Microscopy-Oxford, vol. 225, no. 3, pp. 264–268, 2007.

 BIBLIOGRAPHY

[333] X. Wang, Z. Niu, S. Li, Q. Wang, and X. Li, “Nanomechanical characterization of polyaniline coated tobacco mosaic virus nanotubes,” Journal Biomedical Materials Research A, vol. 87, no. 1, pp. 8–14, 2008.

[334] P. Stockley and R. Twarock, Emerging Topics in Physical Virology. London: Imperial College Press, 2010.

[335] D. E. Smith, S. J. Tans, S. B. Smith, S. Grimes, D. L. Anderson, and C. Bustamante, “The bacteriophage phi 29 portal motor can package DNA against a large internal force,” Nature, vol. 413, no. 6857, pp. 748–752, 2001.

[336] P. K. Purohit, M. M. Inamdar, P. D. Grayson, T. M. Squires, J. Kondev, and R. Phillips, “Forces during bacteriophage DNA packaging and ejection,” Biophysical Journal, vol. 88, no. 2, pp. 851–866, 2005.

[337] E. Nurmemmedov, M. Castelnovo, C. E. Catalano, and A. Evilevitch, “Biophysics of viral infectivity: matching genome length with capsid size,” Quarterly Reviews of Biophysics, vol. 40, no. 4, pp. 327–356, 2007.

[338] T. F. Anderson, C. Rappaport, and N. A. Muscatine, “On the structure and osmotic properties of phage particles,” Annales De L Institut Pasteur, vol. 84, no. 1, pp. 5–15, 1953.

[339] M. Guthold, G. Matthews, A. Negishi, R. M. Taylor, D. Erie, F. P. Brooks, and R. Superﬁne, “Quantitative manipulation of DNA and viruses with the nanomanipulator scanning force microscope,” Surface and Interface Analysis, vol. 27, pp. 437–443, 1999.

[340] C. Carrasco, M. Castellanos, P. J. de Pablo, and M. G. Mateu, “Manipulation of the mechanical properties of a virus by protein engineering,” Proceedings of the National Academy of Sciences of the United States of America, vol. 105, no. 11, pp. 4150–4155, 2008.

[341] D. Q. Zhang, R. Konecny, N. A. Baker, and J. A. McCammon, “Electrostatic interaction between RNA and protein capsid in cowpea chlorotic mottle virus simulated by a coarse-grain rna model and a monte carlo approach,” Biopolymers, vol. 75, no. 4, pp. 325–337, 2004.

[342] W. W. Newcomb and J. C. Brown, “Induced extrusion of dna from the capsid of herpes- simplex virus type-1,” Journal of Virology, vol. 68, no. 1, pp. 433–440, 1994.

[343] F. L. Homa and J. C. Brown, “Capsid assembly and DNA packaging in herpes simplex virus,” Reviews In Medical Virology, vol. 7, no. 2, pp. 107–122, 1997.

[344] J. S. Baines and S. K. Weller, Cleavage and packaging of Herpes Simplex Virus 1 DNA, pp. 135–150. New York: Kluwer Academic/Plenum Publishers, 2005.

 BIBLIOGRAPHY

[345] A. C. Steven, J. B. Heymann, N. Q. Cheng, B. L. Trus, and J. F. Conway, “Virus maturation: dynamics and mechanism of a stabilizing structural transition that leads to infectivity,” Current Opinion in Structural Biology, vol. 15, no. 2, pp. 227–236, 2005.

[346] M. L. Baker, W. Jiang, F. J. Rixon, and W. Chiu, “Common ancestry of herpesviruses and tailed DNA bacteriophages,” Journal of Virology, vol. 79, no. 23, pp. 14967–14970, 2005.

[347] D. J. McGeoch, F. J. Rixon, and A. J. Davison, “Topics in herpes virus genomics and evolution,” Virus Research, vol. 117, no. 1, pp. 90–104, 2006.

[348] W. W. Newcomb, F. L. Homa, D. R. Thomsen, F. P. Booy, B. L. Trus, A. C. Steven, J. V. Spencer, and J. C. Brown, “Assembly of the herpes simplex virus capsid: Char- acterization of intermediates observed during cell-free capsid formation,” Journal of Molecular Biology, vol. 263, no. 3, pp. 432–446, 1996.

[349] B. L. Trus, W. W. Newcomb, N. Q. Cheng, G. Cardone, L. Marekov, F. L. Horna, J. C. Brown, and A. C. Steven, “Allosteric signaling and a nuclear exit strategy: Binding of UL25/UL17 heterodimers to DNA-ﬁlled HSV-1 capsids,” Molecular Cell, vol. 26, no. 4, pp. 479–489, 2007.

[350] G. Sherman and S. L. Bachenheimer, “Characterization of intranuclear capsids made by Ts morphogenic mutants of HSV-1,” Virology, vol. 163, no. 2, pp. 471–480, 1988.

[351] J. M. Fox, X. X. Zhao, J. A. Speir, and M. J. Young, “Analysis of a salt stable mutant of cowpea chlorotic mottle virus,” Virology, vol. 222, no. 1, pp. 115–122, 1996.

[352] T. Douglas, E. Strable, D. Willits, A. Aitouchen, M. Libera, and M. Young, “Protein engineering of a viral cage for constrained nanomaterials synthesis,” Advanced Materi- als, vol. 14, no. 6, pp. 415–418, 2002.

[353] J. Lidmar, L. Mirny, and D. R. Nelson, “Virus shapes and buckling transitions in spherical shells,” Physical Review E, vol. 68, no. 5, p. 051910, 2003.

[354] P. K. Purohit, J. Kondev, and R. Phillips, “Mechanics of DNA packaging in viruses,” Proceedings of the National Academy of Sciences of the United States of America, vol. 100, no. 6, pp. 3173–3178, 2003.

[355] F. Tama and C. L. Brooks, “Diversity and identity of mechanical properties of icosahedral viral capsids studied with elastic network normal mode analysis,” Journal of Molecular Biology, vol. 345, no. 2, pp. 299–314, 2005.

[356] R. Zandi and D. Reguera, “Mechanical properties of viral capsids,” Physical Review E, vol. 72, no. 2, p. 021917, 2005.

[357] G. A. Vliegenthart and G. Gompper, “Mechanical deformation of spherical viruses with icosahedral symmetry,” Biophysical Journal, vol. 91, no. 3, pp. 834–841, 2006.

 BIBLIOGRAPHY

[358] M. M. Gibbons and W. S. Klug, “Nonlinear ﬁnite-element analysis of nanoindentation of viral capsids,” Physical Review E, vol. 75, no. 3, p. 031901, 2007.

[359] Z. Yang, I. Bahar, and M. Widom, “Vibrational dynamics of icosahedrally symmetric biomolecular assemblies compared with predictions based on continuum elasticity,” Biophysical Journal, vol. 96, no. 11, pp. 4438–4448, 2009.

[360] P. L. Freddolino, A. S. Arkhipov, S. B. Larson, A. McPherson, and K. Schulten, “Molec- ular dynamics simulations of the complete satellite tobacco mosaic virus,” Structure, vol. 14, no. 3, pp. 437–449, 2006.

[361] A. Arkhipov, P. L. Freddolino, and K. Schulten, “Stability and dynamics of virus capsids described by coarse-grained modeling,” Structure, vol. 14, no. 12, pp. 1767–1777, 2006.

[362] M. Zink and H. Grubmuller, “Mechanical properties of the icosahedral shell of southern bean mosaic virus: a molecular dynamics study,” Biophysical Journal, vol. 96, no. 4, pp. 1350–63, 2009.

[363] T. Ando, T. Uchihashi, N. Kodera, D. Yamamoto, A. Miyagi, M. Taniguchi, and H. Ya- mashita, “High-speed afm and nano-visualization of biomolecular processes,” Pﬂugers Archiv-European Journal of Physiology, vol. 456, no. 1, pp. 211–225, 2008.

[364] M. Baclayon, G. K. Shoemaker, C. Uetrecht, S. E. Crawford, M. K. Estes, B. V. V. Prasad, A. J. R. Heck, G. J. L. Wuite, and W. H. Roos, “Prestress Strengthens the Shell of Norwalk Virus Nanoparticles,” Nano Letters, vol. 11, no. 11, pp. 4865–4869, 2011.

[365] J. Snijder, I. L. Ivanovska, M. Baclayon, W. H. Roos, and G. J. L. Wuite, “Probing the impact of loading rate on the mechanical properties of viral nanoparticles,” Micron, vol. 43, no. 12, SI, pp. 1343–1350, 2012.

[366] N. F. Steinmetz, E. Bock, R. P. Richter, J. P. Spatz, G. P. Lomonossoﬀ, and D. J. Evans, “Assembly of multilayer arrays of viral nanoparticles via biospeciﬁc recognition: A quartz crystal microbalance with dissipation monitoring study,” Biomacromolecules, vol. 9, no. 2, pp. 456–462, 2008.

[367] F. M. Brunel, J. D. Lewis, G. Destito, N. F. Steinmetz, M. Manchester, H. Stuhlmann, and P. E. Dawson, “Hydrazone ligation strategy to assemble multifunctional viral nanoparticles for cell imaging and tumor targeting,” Nano Letters, vol. 10, no. 3, pp. 1093–1097, 2010.

[368] M. E. Hardy and M. K. Estes, “Completion of the Norwalk virus genome sequence,” Virus Genes, vol. 12, no. 3, pp. 287–290, 1996.

[369] K. Y. Green, J. F. Lew, J. Xi, A. Z. Kapikian, and M. K. Estes, “Comparison of the reactivities of baculovirus-expressed recombinant Norwalk virus capsid antigen with

 BIBLIOGRAPHY

those of the native Norwalk virus-antigen in serologic assays and some epidemiologic observations,” Journal of Clinical Microbiology, vol. 31, no. 8, pp. 2185–2191, 1993.

[370] M. Tan, R. S. Hegde, and X. Jiang, “The P domain of norovirus capsid protein forms dimer and binds to histo-blood group antigen receptors,” Journal of Virology, vol. 78, no. 12, pp. 6233–6242, 2004.

[371] N. Wang, I. M. Tolic-Norrelykke, J. X. Chen, S. M. Mijailovich, J. P. Butler, J. J. Fredberg, and D. Stamenovic, “Cell prestress. i. stiﬀness and prestress are closely associated in adherent contractile cells,” American Journal of Physiology-Cell Physiology, vol. 282, no. 3, pp. C606–C616, 2002.

[372] M. L. Gardel, F. Nakamura, J. H. Hartwig, J. C. Crocker, T. P. Stossel, and D. A. Weitz, “Prestressed F-actin networks cross-linked by hinged ﬁlamins replicate mechanical properties of cells,” Proceedings of the National Academy of Sciences of the United States of America, vol. 103, no. 6, pp. 1762–1767, 2006.

[373] C. Carrasco, A. Luque, M. Hernando-Perez, R. Miranda, J. L. Carrascosa, P. A. Serena, M. de Ridder, A. Raman, J. Gomez-Herrero, I. A. T. Schaap, D. Reguera, and P. J. de Pablo, “Built-in mechanical stress in viral shells,” Biophysical Journal, vol. 100, no. 4, pp. 1100–1108, 2011.

[374] A. Zlotnick, N. Cheng, J. F. Conway, F. P. Booy, A. C. Steven, S. J. Stahl, and P. T. Wingﬁeld, “Dimorphism of hepatitis B virus capsids is strongly inﬂuenced by the C-terminus of the capsid protein,” Biochemistry, vol. 35, no. 23, pp. 7412–7421, 1996.

[375] C. Uetrecht and A. J. R. Heck, “Modern biomolecular mass spectrometry and its role in studying virus structure, dynamics, and assembly,” Angewandte Chemie-International Edition, vol. 50, no. 36, pp. 8248–8262, 2011.

[376] K. Lorenzen, C. Versluis, E. van Duijn, R. H. H. van den Heuvel, and A. J. R. Heck, “Optimizing macromolecular tandem mass spectrometry of large non-covalent complexes using heavy collision gases,” International Journal of Mass Spectrometry, vol. 268, no. 2-3, pp. 198–206, 2007.

[377] F. Krause and H. Seelert, “Detection and analysis of protein-protein interactions of or- ganellar and prokaryotic proteomes by blue native and colorless native gel electrophoresis,” Current Protocols Protein Science, vol. Chapter 19, p. Unit 19 18, 2008.

[378] Y. G. Kuznetsov and A. McPherson, “Atomic force microscopy in imaging of viruses and virus-infected cells,” Microbiology Molecular Biology Reviews, vol. 75, no. 2, pp. 268– 85, 2011.

[379] J. L. Cuellar, F. Meinhoevel, M. Hoehne, and E. Donath, “Size and mechanical stability of Norovirus capsids depend on pH: a nanoindentation study,” Journal of General Virology, vol. 91, pp. 2449–2456, 2010.

 BIBLIOGRAPHY

[380] A. Evilevitch, W. H. Roos, I. L. Ivanovska, M. Jeembaeva, B. Jonsson, and G. J. L. Wuite, “Eﬀects of salts on internal DNA pressure and mechanical properties of phage capsids,” Journal of Molecular Biology, vol. 405, no. 1, pp. 18–23, 2011.

[381] M. Baclayon, P. van Ulsen, S. Abeln, W. H. Roos, and G. J. L. Wuite, “Folding-driven transport mechanism of autotransporters studied through the unfolding pathways of Haemoglobin protease using AFM,” manuscript in preparation, 2013.

[382] I. Henderson, F. Navarro-Garcia, and J. Nataro, “The great escape: structure and function of the autotransporter proteins,” Trends in Microbiology, vol. 6, no. 9, pp. 370– 378, 1998.

[383] M. Desvaux, N. Parham, and I. Henderson, “The autotransporter secretion system,” Research in Microbiology, vol. 155, no. 2, pp. 53–60, 2004.

[384] N. Dautin and H. D. Bernstein, “Protein secretion in gram-negative bacteria via the autotransporter pathway,” Annual Review of Microbiology, vol. 61, pp. 89–112, 2007.

[385] P. van Ulsen, “Protein folding in bacterial adhesion: Secretion and folding of classical monomeric autotransporters,” in Bacterial Adhesion: Chemistry, Biology and Physics (Linke, D and Goldman, A, ed.), vol. 715 of Advances in Experimental Medicine and Biology, pp. 125–142, 2011.

[386] J. Grijpstra, M. P. Bos, and J. Tommassen, “Assembly of bacterial outer membrane proteins,” Methods in molecular biology, vol. 966, pp. 223–37, 2013.

[387] C. Oomen, P. van Ulsen, P. Van Gelder, M. Feijen, J. Tommassen, and P. Gros, “Structure of the translocator domain of a bacterial autotransporter,” EMBO Journal, vol. 23, no. 6, pp. 1257–1266, 2004.

[388] A. Sauri, N. Oreshkova, Z. Soprova, W. S. P. Jong, M. Sani, P. J. Peters, J. Luirink, and P. van Ulsen, “Autotransporter beta-domains have a speciﬁc function in protein secretion beyond outer-membrane targeting,” Journal of Molecular Biology, vol. 412, no. 4, pp. 553–567, 2011.

[389] E. Braselmann and P. L. Clark, “Autotransporters: The cellular environment reshapes a folding mechanism to promote protein transport,” Journal of Physical Chemistry Letters, vol. 3, no. 8, pp. 1063–1071, 2012.

[390] N. Dautin, “Serine Protease Autotransporters of Enterobacteriaceae (SPATEs): Bio- genesis and Function,” Toxins, vol. 2, no. 6, pp. 1179–1206, 2010.

[391] J. H. Peterson, P. Tian, R. Ieva, N. Dautin, and H. D. Bernstein, “Secretion of a bacterial virulence factor is driven by the folding of a C-terminal segment,” Proceedings of the National Academy of Sciences of the United States of America, vol. 107, no. 41, pp. 17739–17744, 2010.

 BIBLIOGRAPHY

[392] J. P. Renn, M. Junker, R. N. Besingi, E. Braselmann, and P. L. Clark, “ATP- independent control of autotransporter virulence protein transport via the folding properties of the secreted protein,” Chemistry & Biology, vol. 19, no. 2, pp. 287–296, 2012.

[393] D. L. Leyton, A. E. Rossiter, and I. R. Henderson, “From self suﬃciency to dependence: mechanisms and factors important for autotransporter biogenesis,” Nature Reviews Microbiology, vol. 10, no. 3, pp. 213–225, 2012.

[394] B. van den Berg, “Crystal Structure of a Full-Length Autotransporter,” Journal of Molecular Biology, vol. 396, no. 3, pp. 627–633, 2010.

[395] A. Sauri, C. M. ten Hagen-Jongman, P. van Ulsen, and J. Luirink, “Estimating the size of the active translocation pore of an autotransporter,” Journal of Molecular Biology, vol. 416, no. 3, pp. 335–345, 2012.

[396] J. P. Renn and P. L. Clark, “A conserved stable core structure in the passenger domain beta-helix of autotransporter virulence proteins,” Biopolymers, vol. 89, no. 5, pp. 420– 427, 2008.

[397] J. P. Renn and P. L. Clark, “Disulﬁde bond-mediated passenger domain stalling as a structural probe of autotransporter outer membrane secretion in vivo,” in Methods in Enzymology: Biothermodynamics Vol 492 (Johnson, ML and Holt, JM and Ackers, GK, ed.), vol. 492 of Methods in Enzymology, pp. 233–251, 2011.

[398] Z. Soprova, A. Sauri, P. van Ulsen, J. R. H. Tame, T. den Blaauwen, W. S. P. Jong, and J. Luirink, “A conserved aromatic residue in the autochaperone domain of the autotransporter hbp is critical for initiation of outer membrane translocation,” Journal of Biological Chemistry, vol. 285, no. 49, pp. 38224–38233, 2010.

[399] A. Sauri, Z. Soprova, D. Wickstrom, J.-W. de Gier, R. C. Van der Schors, A. B. Smit, W. S. P. Jong, and J. Luirink, “The Bam (Omp85) complex is involved in secretion of the autotransporter haemoglobin protease,” Microbiology, vol. 155, no. 12, SI, pp. 3982– 3991, 2009.

[400] K. Nishimura, N. Tajima, Y.-H. Yoon, S.-Y. Park, and J. R. H. Tame, “Autotrans- porter passenger proteins: virulence factors with common structural themes,” Journal of Molecular Medicine, vol. 88, no. 5, pp. 451–458, 2010.

[401] P. Emsley, I. Charles, N. Fairweather, and N. Isaacs, “Structure of Bordetella pertussis virulence factor P.69 pertactin,” Nature, vol. 381, no. 6577, pp. 90–92, 1996.

[402] C. Eslava, F. Navarro-Garcia, J. Czeczulin, I. Henderson, A. Cravioto, and J. Nataro, “Pet, an autotransporter enterotoxin from enteroaggregative Escherichia coli,” Infection and Immunity, vol. 66, no. 7, pp. 3155–3163, 1998.

 BIBLIOGRAPHY

[403] T. A. Johnson, J. Qiu, A. G. Plaut, and T. Holyoak, “Active-Site Gating Regu- lates Substrate Selectivity in a Chymotrypsin-Like Serine Protease: The Structure of Haemophilus inﬂuenzae Immunoglobulin A1 Protease,” Journal of Molecular Biology, vol. 389, no. 3, pp. 559–574, 2009.

[404] B. Otto, R. Sijbrandi, J. Luirink, B. Oudega, J. Heddle, K. Mizutani, S. Park, and J. Tame, “Crystal structure of hemoglobin protease, a heme binding autotransporter protein from pathogenic Escherichia coli,” Journal of Biological Chemistry, vol. 280, no. 17, pp. 17339–17345, 2005.

[405] M. Junker, C. Schuster, A. McDonnell, K. Sorg, M. Finn, B. Berger, and P. Clark, “Pertactin beta-helix folding mechanism suggests common themes for the secretion and folding of autotransporter proteins,” Proceedings of the National Academy of Sciences of the United States of America, vol. 103, no. 13, pp. 4918–4923, 2006.

[406] Y. T. Yen, M. Kostakioti, I. R. Henderson, and C. Stathopoulos, “Common themes and variations in serine protease autotransporters,” Trends in Microbiology, vol. 16, no. 8, pp. 370–379, 2008.

[407] B. Otto, S. van Dooren, J. Nuijens, J. Luirink, and B. Oudega, “Characterization of a hemoglobin protease secreted by the pathogenic Escherichia coli strain EB1,” Journal of Experimental Medicine, vol. 188, no. 6, pp. 1091–1103, 1998.

[408] N. Tajima, F. Kawai, S.-Y. Park, and J. R. H. Tame, “A Novel Intein-Like Autoprote- olytic Mechanism in Autotransporter Proteins,” Journal of Molecular Biology, vol. 402, no. 4, pp. 645–656, 2010.

[409] W. S. P. Jong, Z. Soprova, K. de Punder, C. M. ten Hagen-Jongman, S. Wagner, D. Wickstrom, J.-W. de Gier, P. Andersen, N. N. van der Wel, and J. Luirink, “A structurally informed autotransporter platform for eﬃcient heterologous protein secretion and display,” Microbial Cell Factories, vol. 11, 2012.

[410] Y. R. Sevastsyanovich, D. L. Leyton, T. J. Wells, C. A. Wardius, K. Tveen-Jensen, F. C. Morris, T. J. Knowles, A. F. Cunningham, J. A. Cole, and I. R. Henderson, “A generalised module for the selective extracellular accumulation of recombinant proteins,” Microbial Cell Factories, vol. 11, 2012.

[411] E. van Bloois, R. T. Winter, H. Kolmar, and M. W. Fraaije, “Decorating microbes: surface display of proteins on Escherichia coli,” Trends in Biotechnology, vol. 29, no. 2, SI, pp. 79–86, 2011.

[412] W. S. P. Jong, A. Sauri, and J. Luirink, “Extracellular production of recombinant proteins using bacterial autotransporters,” Current Opinion in Biotechnology, vol. 21, no. 5, pp. 646–652, 2010.

 BIBLIOGRAPHY

[413] J. Jose and T. F. Meyer, “The autodisplay story, from discovery to biotechnical and biomedical applications,” Microbiology and Molecular Biology Reviews, vol. 71, no. 4, pp. 600+, 2007.

[414] W. S. P. Jong, C. M. ten Hagen-Jongman, T. den Blaauwen, D. J. Slotboom, J. R. H. Tame, D. Wickstrom, J.-W. de Gier, B. R. Otto, and J. Luirink, “Limited tolerance towards folded elements during secretion of the autotransporter Hbp,” Molecular Mi- crobiology, vol. 63, no. 5, pp. 1524–1536, 2007.

[415] W. Kalle and P. Strappe, “Atomic force microscopy on chromosomes, chromatin and DNA: A review,” Micron, vol. 43, no. 12, SI, pp. 1224–1231, 2012.

[416] D. J. Billingsley, W. A. Bonass, N. Crampton, J. Kirkham, and N. H. Thomson, “Single-molecule studies of DNA transcription using atomic force microscopy,” Physical Biology, vol. 9, no. 2, 2012.

[417] C.-A. Lamontagne, C. M. Cuerrier, and M. Grandbois, “AFM as a tool to probe and manipulate cellular processes,” European Journal of Physiology, vol. 456, no. 1, pp. 61– 70, 2008.

[418] A. Ikai and R. Afrin, “Toward mechanical manipulations of cell membranes and membrane proteins using an atomic force microscope - An invited review,” Cell Biochemistry and Biophysics, vol. 39, no. 3, pp. 257–277, 2003.

[419] S. Kumar and J. Hoh, “Probing the machinery of intracellular traﬃcking with the atomic force microscope,” Traﬃc, vol. 2, no. 11, pp. 746–756, 2001.

[420] S. Allen, X. Chen, J. Davies, M. Davies, A. Dawkes, J. Edwards, C. Roberts, J. Sefton, S. Tendler, and P. Williams, “Detection of antigen-antibody binding events with the atomic force microscope,” Biochemistry, vol. 36, no. 24, pp. 7457–7463, 1997.

[421] A. E. X. Brown, R. I. Litvinov, D. E. Discher, and J. W. Weisel, “Forced unfolding of coiled-coils in ﬁbrinogen by single-molecule AFM,” Biophysical Journal, vol. 92, no. 5, pp. L39–L41, 2007.

[422] G. Lee, K. Abdi, Y. Jiang, P. Michaely, V. Bennett, and P. Marszalek, “Nanospring behaviour of ankyrin repeats,” Nature, vol. 440, no. 7081, pp. 246–249, 2006.

[423] S. Kumar and M. S. Li, “Biomolecules under mechanical force,” Physics Reports - Review Section of Physics Letters, vol. 486, no. 1-2, pp. 1–74, 2010.

[424] J. R. Forman and J. Clarke, “Mechanical unfolding of proteins: insights into biology, structure and folding,” Current Opinion in Structural Biology, vol. 17, no. 1, pp. 58–66, 2007.

 BIBLIOGRAPHY

[425] V. E. V. Ferrero, L. Andolﬁ, G. Di Nardo, S. J. Sadeghi, A. Fantuzzi, S. Cannistraro, and G. Gilardi, “Protein and Electrode Engineering for the Covalent Immobilization of P450 BMP on Gold,” Analytical Chemistry, vol. 80, no. 22, pp. 8438–8446, 2008.

[426] H. Dietz and M. Rief, “Detecting molecular ﬁngerprints in single molecule force spectroscopy using pattern recognition,” Japanese Journal of Applied Physics, vol. 46, no. 8B, pp. 5540–5542, 2007.

[427] E. M. Puchner, G. Franzen, M. Gautel, and H. E. Gaub, “Comparing proteins by their unfolding pattern,” Biophysical Journal, vol. 95, no. 1, pp. 426–434, 2008.

[428] A. Noy, “Force spectroscopy 101: how to design, perform, and analyze an AFM-based single molecule force spectroscopy experiment,” Current Opinion in Chemical Biology, vol. 15, no. 5, pp. 710–718, 2011.

[429] D. B. Staple, S. H. Payne, A. L. C. Reddin, and H. J. Kreuzer, “Stretching and unfolding of multidomain biopolymers: a statistical mechanics theory of titin,” Physical Biology, vol. 6, no. 2, 2009.

[430] W. Kabsch and C. Sander, “Dictionary of Protein Secondary Structure: Pattern Recog- nition of Hydrogen-Bonded and Geometrical Features,” Biopolymers, vol. 22, no. 12, pp. 2577–2637, 1983.

[431] C. Chyan, F. Lin, H. Peng, J. Yuan, C. Chang, S. Lin, and G. Yangy, “Reversible mechanical unfolding of single ubiquitin molecules,” Biophysical Journal, vol. 87, no. 6, pp. 3995–4006, 2004.

[432] M. Mickler, R. I. Dima, H. Dietz, C. Hyeon, D. Thirumalai, and M. Rief, “Revealing the bifurcation in the unfolding pathways of GFP by using single-molecule experiments and simulations,” Proceedings of the National Academy of Sciences of the United States of America, vol. 104, no. 51, pp. 20268–20273, 2007.

[433] J. Stigler, F. Ziegler, A. Gieseke, J. C. M. Gebhardt, and M. Rief, “The Complex Folding Network of Single Calmodulin Molecules,” Science, vol. 334, no. 6055, pp. 512– 516, 2011.

[434] E. Evans, “Probing the relation between Force - Lifetime - and Chemistry in single molecular bonds,” Annual Review of Biophysics and Biomolecular Structure, vol. 30, pp. 105–128, 2001.

[435] O. K. Dudko, G. Hummer, and A. Szabo, “Theory, analysis, and interpretation of single-molecule force spectroscopy experiments,” Proceedings of the National Academy of Sciences of the United States of America, vol. 105, no. 41, pp. 15755–15760, 2008.

[436] Y. Oberdorfer, H. Fuchs, and A. Janshoﬀ, “Conformational analysis of native ﬁ- bronectin by means of force spectroscopy,” Langmuir, vol. 16, no. 26, pp. 9955–9958, 2000.

 BIBLIOGRAPHY

[437] M. Yoder, N. Keen, and F. Jurnak, “New Domain Motif - The Structure of Pectate Lyase-C, a Secreted Plant Virulence Factor,” Science, vol. 260, no. 5113, pp. 1503–1507, 1993.

[438] F. Jurnak, M. Yoder, R. Pickersgill, and J. Jenkins, “Parallel Beta Domains - A New Fold in Protein Structures,” Current Opinion in Structural Biology, vol. 4, no. 6, pp. 802–806, 1994.

[439] S. Steinbacher, R. Seckler, S. Miller, B. Steipe, R. Huber, and P. Reinemer, “Crystal structure of P22 tailspike protein - interdigitated subunits in a thermostable trimer,” Science, vol. 265, no. 5170, pp. 383–386, 1994.

[440] J. Jenkins and R. Pickersgill, “The architecture of parallel beta-helices and related folds,” Progress in Biophysics & Molecular Biology, vol. 77, no. 2, pp. 111–175, 2001.

[441] G. McGaughey, M. Gagne, and A. Rappe, “Pi-Stacking Interactions - Alive and Well in Proteins,” Journal of Biological Chemistry, vol. 273, no. 25, pp. 15458–15463, 1998.

[442] M. Waters, “Aromatic interactions in model systems,” Current Opinion in Chemical Biology, vol. 6, no. 6, pp. 736–741, 2002.

[443] N. Kannan and S. Vishveshwara, “Aromatic clusters: a determinant of thermal stability of thermophilic proteins,” Protein Engineering, vol. 13, no. 11, pp. 753–761, 2000.

[444] O. Makin, E. Atkins, P. Sikorski, J. Johansson, and L. Serpell, “Molecular basis for amyloid ﬁbril formation and stability,” Proceedings of the National Academy of Sci- ences of the United States of America, vol. 102, no. 2, pp. 315–320, 2005.

[445] F. Chiti and C. M. Dobson, “Protein misfolding, functional amyloid, and human disease,” Annual Review of Biochemistry, vol. 75, pp. 333–366, 2006.

[446] R. Simkovsky and J. King, “An elongated spine of buried core residues necessary for in vivo folding of the parallel beta-helix of P22 tailspike adhesin,” Proceedings of the Na- tional Academy of Sciences of the United States of America, vol. 103, no. 10, pp. 3575– 3580, 2006.



List of Publications

The following publications are reproduced as chapters in this thesis:

Chapter 2: M. Baclayon, W.H. Roos, and G.J.L. Wuite, ‘Sampling protein form and function with the atomic force microscope’. Molecular & Cellular Proteomics 9, 8, pp.1678-1688, 2010.

Chapter 3: G. Farge, M. Mehmedovic, M. Baclayon, S.M. van den Wildenberg, W.H. Roos, C.M. Gustafsson, G.J.L. Wuite, and M. Falkenberg, ‘Nucleoid packaging blocks replication and transcription of mitochondrial DNA’, manuscript in preparation, 2013.

D.A. Heuveling, G.W.M. Visser, M. Baclayon, W.H. Roos, G.J.L. Wuite, O.S. Hoekstra, C.R. Leemans, R. de Bree, and G.A.M.S. van Dongen, ‘89Zr Nanocol- loidal albumin-based PET/CT lymphoscintigraphy for sentinel node detection in head and neck cancer: Preclinical results’, Journal of Nuclear Medicine 52, 10, pp.1580-1584, 2011.

D.A. Heuveling, G.W.M. Visser, M. de Groot, J.F. de Boer, M. Baclayon, W.H. Roos, G.J.L. Wuite, C.R. Leemans, R. de Bree, and G.A.M.S. van Dongen, ‘Nanocolloidal albumin-IRDye 800CW: a near-infrared ﬂuorescent tracer with optimal retention in the sentinel lymph node’, European Journal of Nuclear Medicine and Molecular Imaging 39, 7, pp.1161-1168, 2012.

G.K. Shoemaker, E. van Duijn, S.E. Crawford, C. Uetrecht, M. Baclayon, W.H. Roos, G.J.L. Wuite, M.K. Estes, B.V.V. Prasad, and A.J.R. Heck, ‘Norwalk virus assembly and stability monitored by mass spectrometry’, Molecular & Cellular Proteomics 9, 8, pp.1742-1751, 2010.

Chapter 4: M. Baclayon, G.J.L. Wuite, and W.H. Roos, ‘Imaging and manipulation of single viruses by atomic force microscopy’, Soft Matter 6, 21, pp.5273-5285, 2010. Chapter 5: M. Baclayon, G.K. Shoemaker, C. Uetrecht, S.E. Crawford, M.K. Estes, B.V.V. Prasad, A.J.R. Heck, G.J.L. Wuite, and W.H. Roos, ‘Prestress strengthens the shell of Norwalk virus nanoparticles’, Nano Letters 11, 11, pp.4865-4869, 2011.

J. Snijder, I.L. Ivanovska, M. Baclayon, W.H. Roos, and G.J.L. Wuite, ‘Probing the impact of loading rate on the mechanical properties of viral nanoparticles’, Micron 43, 12, pp.1343-1350, 2012.

Chapter 6: M. Baclayon, P. van Ulsen, S. Abeln, W.H. Roos, and G.J.L. Wuite, ‘Folding-driven secretion mechanism of autotransporters studied through the unfolding pathways of Haemoglobin protease using AFM’, manuscript in preparation, 2013.

 Summary

Elasticity of Biomolecules: probing, pushing and pulling using atomic force microscopy

The elasticity of biomolecules from DNA, proteins, colloids to viruses were studied using the combined imaging and force spectroscopy capabilities of atomic force microscopy (AFM). The AFM is similar to a blind man’s stick. It creates a topographical picture of the surface by mechanical probing using a flexible cantilever (as the stick) with a very sharp tip. Using high-precision piezo-electric scanners, it can image biomolecules with sub-nanometer resolution (1/1000 000 000 of a meter). As a contact microscope, it is also used to apply forces by pushing and pulling on the sample and measure forces by monitoring the amount of cantilever bending. Using very flexible cantilevers, the AFM can apply and detect forces with a few piconewton resolution (1/1000 000 000 000 of a newton). As an imaging technique, I used the AFM to investigate how the compaction of DNA by a protein called TFAM affects the replication/transcription processes in the mitochondria, how radioactive and fluorescent labeling affects the size and uptake of nanocolloidal albumins in detecting tumors and how capsid proteins of Norovirus spontaneously disassemble and reassemble into virus-like particles. From the conformation images of DNA molecules at different TFAM concentrations, it is found that the physiological concentration of TFAM is a narrow window in which both nearly-naked and compacted DNA molecules exist, suggesting that TFAM acts as a regulation factor in controlling the number of mitochondrial DNA molecules available for replication and transcription. The AFM images show that the labeling of nanocolloidal albumin with 99mTc, 89Zr and IRDye 800CW dye does not affect the size and uptake of these particles by cancer cells. The confirmation of these factors opens up new techniques such as PET scan and fluorescence microscopy for tumor imaging and diagnostics. AFM imaging of Norovirus-like particles show that they reversibly disassemble and reassemble into 38-nm particles with T=3 icosahedral symmetry at acidic conditions (2

 Nederlandse Samenvatting

Elasticiteit van biomoleculen: voelen, duwen en trekken met de atoomkrachtmicroscoop

De elasticiteit van biomoleculen zoals DNA, eiwitten, colloïdes en virussen zijn bestudeerd met een atoomkrachtmicroscoop (AFM, van het engelse atomic force microscope): een combinatie van beeldvorming met krachtspectroscopie. De AFM kan worden vergeleken met een blindenstok: door met een flexibele hefboom met een hele scherpe punt het oppervlakte af te tasten kan een topografische afbeelding van het oppervlakte gemaakt worden. Met zeer nauwkeurige piëzo-electrische scanners kun- nen zo biomoleculen afgebeeld worden met een resolutie van beter dan een nanometer (1/1000 000 000e van een meter). Als contactmicroscoop kan de AFM ook gebruikt worden om kracht uit te oefenen door op een monster te drukken of eraan te trekken. Tegelijkertijd kan de kracht gemeten worden door de afbuiging van de hefboom te meten. Door zeer flexibele hefbomen te gebruiken, kan met de AFM een kracht van een paar piconewton (1/1 000 000 000 000e van een newton) uitgeoefend en gede- tecteerd worden. Ik heb de AFM als beeldvormingstechniek gebruikt om te bestuderen hoe DNA gecompacteerd wordt door TFAM, een eiwit dat replicatie en transcriptie in mi- tochondriën beïnvloedt, ik heb gekeken hoe radioactieve en fluorescente labels het formaat en de opname van nanocolloïdische albumines beïnvloeden in het detecteren van tumoren en ik heb gekeken hoe de schaal van het Norovirus spontaan uiteenvalt en weer assembleert in virusachtige deeltjes. Uit de afbeeldingen van DNA-moleculen bij verschillende TFAM-concentraties hebben we afgeleid dat de fysiologische concentratie van TFAM binnen een smalle band valt waarbij zowel bijna naakte als gecompacteerde DNA moleculen bestaan. Dit suggereert dat TFAM zich gedraagt als een regulatiefactor door het aantal mitochon- driale DNA-moleculen dat beschikbaar is voor regulatie en transcriptie te controleren. De AFM-afbeeldingen laten zien dat de labeling van nanocolloïdische albumines met 99mTc, 89Zr en IRDye 800CW het formaat en de opname van de bolletjes door kanker- cellen niet beïnvloedt. Deze bevestiging opent de weg naar nieuwe technieken zoals PET-scan en fluorescentiemicroscopie voor beeldvorming en diagnosticering van tumoren. AFM-afbeeldingen van Norovirusachtige deeltjes laten zien dat deze deeltjes in een zuur milieu (2

 Acknowledgments

The completion of my PhD would have not been realized without the contributions and support that I received from a lot of people. I would like to take this opportunity to express my endless gratitude to them. I would like to thank Gijs for giving me the opportunity to be part of the CoSy group and do research at the forefront of single-molecule biophysics. I appreciate the balance of independence and direction you gave me to pursue projects that I find interesting. You have demanded excellence in research, but you are also very generous with expressing your enthusiasm and appreciation. That gave me boosts of confidence and encouragement on several occasions and have kept me going despite the challenges. My gratitude to Wouter for introducing me to atomic force microscopy and the nanoindentation experiments on viruses. I appreciate very much your principal contributions to the reviews we have written and most especially how you have rounded up the Norovirus project to a very nice publication. I also thank you for imparting helpful feedbacks on the different research projects I have undertaken. Most of the research projects would have not been made possible without the collaborations I had with scientists from different groups. The Norovirus project is a collaboration with Utrecht University and Baylor College of Medicine. I would like to specially thank Glen Shoemaker. It was a pleasure to work with you and I enjoyed our conversations about anything from mass spectrometry to Canada. I would also like to thank Charlotte Uetrecht, Joost Snijder, Albert Heck and Ventakaram Prasad for your feedbacks on the paper. The Nanocoll project is a collaboration with the VU medical center. My thanks to Derreck Heuveling for the good and interesting collaboration. I am very thankful to Geraldine Farge whom I have worked with on the mitochondrial DNA-TFAM and POLRMT projects. I really enjoyed working with you and I really appreciate your friendship and our morning coffee breaks and chitchats. I am very happy that the work we did on DNA-TFAM complexes formed part of a nice publication. Unfortunately, the POLRMT project did not advance as we would have liked it to, but I have learned a lot that I do not regret us going for it despite the challenges. My special thanks to Peter van Ulsen for the great collaboration on autotransporters. The folding-driven hypothesis for the secretion of Hbp is one of the most exciting and interesting research questions I have come across. Your enthusiasm about Hbp is very infectious and I really appreciate your words of encouragement during the difficult times. I would also like to thank Sanne Abeln for joining the team and for making sense of our pulling data. The unfolding experiments were no less than challenging, but I consider myself very lucky to be surrounded and to work with brilliant and accomplished scientists. I am looking forward to seeing our hard work completed in a nice publication. I am grateful to Jan Rector and Martin Slaman for assisting me in preparing gold-coated substrates that were indispensable for my pulling experiments. Thank you very much for your patience and for your time in training me how to use the sputtering machine. I would like to express my gratitude and great appreciation to my CoSy family. I thank you all for the camaraderie you have extended to me for four and a half years at the VU. I would always have fun and fond memories of our lunches together, the afternoon tea breaks with delicious Dutch cookies and cakes, and the group outings and parties. The time I spent in the VU will always remain with me and I will surely miss you all. I thank the CoSy girls for the wonderful all-girls dinner together. You are a bunch of good-natured and fun-loving women: Ineke, Jona, Els, Astrid, Geraldine, Steph, Ienas, Seyda, Miriam, Sandrijn, Claudia, Sandrine, Kelly, Stefanie, Myrthe, Vandana, Mariska, Karsta, Yan; I thank you all for your friendship and I wish you success in your future endeavors. I am looking forward to joining the girls’ dinner even if I’m not with the VU anymore. I would like to thank the gentlemen: Erwin, Daan, Aravindan, Onno, Niels, Bram, Andrea, Gerrit, Siet, Tjalle, Felix, Peter, Pierre, Graeme, Jules, Andreas, Iddo, Jordi, Douwe, Tim, Johannes, Jianan, Joshua, Matthijs, Jelmer, Frank; your antics and playfulness have made my days in the VU wonderful. I wish you all success in your careers and future endeavors. Special thanks to Ineke for the Dutch translation of the summary. Bedankt voor alles, Ineke! Ik vind het heel mooi. I would like to thank our Filipino circle here in the Netherlands. The Filipinos I call friends are strangely not in the Philippines but are all here in the Netherlands. Joy, Marc, Albert, Jane, Ado, Yoli, Onats, (with our little prince, Julian), Tessa, Ronald, Susan, Carmi, Merced, the honorary Filipina Min, Ate Hermie, the honorary Filipino Piet; your great company has made Edcel and I feel less homesick. A special

 6 Acknowledgments thanks also to a very special family in Germany: Rose Jean, Olaf and Sandy. Thank you for the visits, parties, videoke, poker and food (and bring-home food, too)! I would like to express my gratitude to Fr. Herman van Engelen for the wonderful and inspiring messages and emails, and most of all for your being thoughtful to Filipinos and Filipino students. I am thankful to May Ann & Gerrit Kuik, Daday & Ed van Berg and their families for inviting us over and making us feel at home here in the Netherlands. To my family back home: Mama Myrna, Papa Nador and sister Nemia; thank you very much for the love and support. I have been away from all of you during my PhD. I miss you and I thank you for always being there for me. I specially dedicate this book to you. Lastly, to Edcel - my partner, my adviser, my critic, my friend; thank you for the love and understanding. I would have not survived all the challenges and diﬃculties without you. The triumph and success have meaning only because I share them with you.

Marian Amsterdam, November 2013

