A general property of disordered proteins is their structural expansion that results in a high macromolecular flexibility. Quasielastic incoherent neutron scattering (QENS) is a well-suited experimental method to study protein dynamics from the picosecond to several nanoseconds and in the Ångström length-scale. In QENS experiments of protein solutions hydrogens act as reporters for the motions of methyl groups or amino acids to which they are bound. Neutron Spin-Echo spectroscopy (NSE) on the other hand, offers the highest energy resolution in the field of neutron spectroscopy and allows the study of slow collective motions in proteins up to several hundred nanoseconds and in the nanometer length-scale. In my presentation, I will summarize recent QENS and NSE results on the dynamics of the intrinsically disordered myelin basic protein (MBP) and the chemically denatured bovine serum albumin (BSA) (1,2,3). Using NSE experiments, we observed a high internal flexibility of the intrinsically disordered MBP and the denatured BSA in addition to centre-of-mass diffusion detected by dynamic light scattering. Internal motions measured by NSE were described using concepts based on polymer theory. The contribution of residue-solvent friction was accounted for using the Zimm model including internal friction (ZIF). Disulphide bonds forming loops of amino acids of the peptide backbone have a major impact on internal dynamics that can be interpreted with a reduced set of Zimm modes.
1. Stadler et al. 2014, Journal of the American Chemical Society 136 (19), 6987-6994
2. Ameseder et al. 2018, Physical Chemistry Chemical Physics 20 (7), 5128-5139
3. Ameseder et al. 2018, The Journal of Physical Chemistry Letters 9, 2469-2473
Protein performs its biological functions by interacting with other proteins. Protein complexes, which are formed as a result of these interactions, consist of two or more components that associate along specific pathways - protein association pathways. The association pathway from monomer to oligomer is critical in a range of biological processes and thus it is of a vital importance to elucidate both atomic-resolution structures of intermediates along the pathway as well as the structure of the final state. Although considerable progress has been made in using experimental and computational techniques to determine start and final structural states, we have a limited understanding of what happens in between.
By enabling both time resolution and structural detail Time-Resolved Small Angle X-ray/Neutron Scattering (TR-SAXS/TR-SANS) is uniquely suited to interrogate complex self-assembly reactions and to provide a molecular understanding of self-assembly pathways. However, the analysis of such data is complicated because scattering arises from a mixture of many components, the information content in each spectrum is limited and there is no framework for simultaneous analysis of data from different data sources. The similar problem is faced when resolving conformational ensembles from small angle scattering data.
To overcome this problem we develop a method that combines a computational structural modeling (which delivers atomic-resolution structures) with experimental data (which provides information about the population of different states). The method applies Bayesian probabilistic model to analyze scattering data from mixtures of oligomeric species, allows for a modeling large structural ensembles, can be used to assess uncertainty of all parameters and minimizes over-fitting. Our software is developed to meet high software standards and will become available to ESS users from the early stage of operation.
To obtain a molecular understanding of IDPs, a combined approach of experiments and simulations is useful. Recently we have shown that a coarse-grained model based on the primitive model, that has been used for modelling polyelectrolytes for over 30 years, works well for a range of IDPs where electrostatics governs the intra- and intermolecular interactions . However, some IDPs have the tendency to self-associate into oligomers, also known as micelles. To simulate the self-associating proteins, further development of the model is performed, using the saliva protein Statherin as model system. For this purpose, Statherin has been characterized by small-angle x-ray scattering and the effect of protein concentration, salt and temperature have been investigated. Preliminary Monte Carlo simulation results follow the experimental trends at lower protein concentrations and provide further insight into shape and polydispersity.
 C. Cragnell, E. Rieloff and M. Skepö, J. Mol. Biol., 2018, https://doi.org/10.1016/j.jmb.2018.03.006
Proteins are the most versatile constituents of the molecular machinery of life, and it is becoming increasingly clear that many of them perform essential functions even though they lack a well-defined folded structure. Single-molecule spectroscopy and fluorescence correlation spectroscopy provide an opportunity for investigating the molecular dynamics of these intrinsically disordered proteins on nanometer length scales and across twelve orders of magnitude in time, even in complex environments, including live cells. A physical description of their behavior is becoming increasingly accessible via the synergy of experiment with theory and simulations.
Over the last two decades, the classical structure-function paradigm has gradually been revisited with the discovery and the increasingly recognized importance of intrinsically disordered proteins (IDPs). IDPs do not rely on a well-defined three-dimensional structure to be functional, but rather exploit their intrinsic conformational dynamics for carrying out a wide range of biological functions. It is estimated that around 40% of the human proteome is intrinsically disordered or contain disordered regions of significant length, and it has been shown that intrinsic disorder is particularly abundant in proteins implicated in human diseases underlining the importance of understanding the conformational properties and functional interactions of IDPs at the molecular level.
Nuclear magnetic resonance (NMR) spectroscopy is the most promising technique for visualizing the structure, dynamics and interactions of IDPs at atomic resolution. Here, our sample-and-select approach will be presented for obtaining representative ensemble descriptions of IDPs on the basis of experimental NMR data providing detailed insight into the conformational sampling of IDPs at amino acid resolution . In addition, experimental NMR approaches will be presented for characterizing the structure, dynamics and kinetics of complexes involving IDPs. Examples will be given of functional protein disorder in important biological systems such as paramyxoviruses , the nuclear pore complex  and cell signaling cascades [4,5].
 Jensen et al, Chem. Rev. 114 (2014) 6632–6660.
 Schneider et al, J. Am. Chem. Soc. 137 (2015) 1220–1229.
 Milles et al, Cell 163 (2015) 734–745.
 Kragelj et al, Proc. Natl. Acad. Sci. U. S. A. 112 (2015) 3409–3414.
 Delaforge et al, J. Am. Chem. Soc. 140 (2018) 1148–1158.
The degree of compaction of the polypeptide chain is a property of key importance in intrinsically disordered proteins. Experimentally, compaction is often measured either by NMR (as the hydrodynamic radius) or SAXS (as the radius of gyration), and more detailed information may also be obtained by NMR paramagnetic relaxation enhancement measurements. A more detailed, atomic-level description of the structure, dynamics and compaction of IDPs, may however, be obtained from molecular simulations, but these need to be validated or generated by comparison with experiments. The comparison between computed conformational ensembles and NMR and SAXS experiments is, however, not trivial. I will discuss recent results that provide new insights into how we can compare atomic level structures of IDPs with NMR diffusion, NMR paramagnetic relaxation enhancement and SAXS experiments. Further, I will discuss recent theoretical and practical progress in using experimental NMR and SAXS data to refine computational models of biomolecules.
Large macromolecular machines, such as proteins and their complexes, are typically very flexible at physiological conditions. Computationally, this flexibility can be approximated with just a few collective molecular motions, computed e.g. using the Normal Mode Analysis (NMA). NMA determines low-frequency motions at a very low cost and these are particularly interesting to the structural biology community.
We have recently introduced a new conceptually simple and computationally efficient method for nonlinear normal mode analysis called NOLB . Overall, the NOLB method produces structures with a better local geometry compared to the standard techniques, especially at large deformation amplitudes, and it also predicts better structural transitions. Finally, the NOLB method is scalable and robust, it typically runs at interactive time rates, and can be applied to very large molecular systems, such as ribosomes.
NMA can be combined with other computational techniques for various applications. I will specifically highlight our flexible fitting methods for small-angle X-ray (SAXS) and neutron (SANS) profiles. This was made possible thanks to our SAXS and SANS packages called Pepsi-SAXS , and Pepsi-SANS . Pepsi-SAXS is a novel and very efficient method that calculates SAXS profiles from atomistic models. It is based on the multipole expansion scheme and is significantly faster with the same level of precision compared to CRYSOL, FoXS and other methods. Recently, we designed a computational scheme that uses the NOLB modes as a low-dimensional representation of the protein motion subspace and optimises protein structures guided by the SAXS and SANS profiles. Overall, this scheme allows to significantly improve the goodness of fit to experimental profiles in a very reasonable computational time.
The biological function of proteins is often related to large-scale domain motions, which are induced or suppressed by the binding of a substrate or due to cosolvents. Domain motions can be related to soft hinges, flexible linker regions or -as in the case of intrinsic unfolded proteins- be native to the unfolded protein structure. These large-scale domain motions in solution cannot be observed by X-ray crystallography or NMR spectroscopy. Small angle scattering by X-rays or neutrons in combination with neutron spin echo spectroscopy (NSE) in solution can be used to observe configurational changes and equilibrium dynamics between functional domains on 1-100 nanosecond timescale.
I present here examples for different types of motions related to the structure of proteins and bioconjugates. Thermal unfolded Ribonuclease A shows polymer like dynamics despite the 4 disulfide bonds restricting the degrees of freedom. Phosphoglycerate kinase shows a clear hinge motion between the main domains. PEGylation seems not to influence domain motion but adds additional internal dynamics in the protein-polymer complex. Immunoglobulin 1 (IGG1) presents a strong dynamics due to the short linkers connecting the Fc with the Fab domains.
Relevant forces and friction will be discussed in terms of the Ornstein-Uhlenbeck process.
(1) Inoue, R.; Biehl, R.; Rosenkranz, T.; Fitter, J.; Monkenbusch, M.; Radulescu, A.; Farago, B.; Richter, D. Biophys J 2010, 99 (7), 2309.
(2) Ciepluch K., Radulescu, A., Hoffmann I., Raba, A., Allgaier, J., Richter D., Biehl R., in review
(3) Stingaciu, L. R.; Ivanova, O.; Ohl, M.; Biehl, R.; Richter, D. Sci. Rep. 2016, 6, 22148.
Protein-protein interactions can influence a range of material properties and dynamic/kinetic behaviors, from aggregation kinetics to solution viscosity, self-association, and solubility. This presentation focuses on dilute and concentrated solutions of monoclonal antibodies and synthetic antibodies, from the perspective of predicting the physical properties and/or behavior of these systems as a function of typical formulation variables (e.g., pH, ionic strength). The results illustrate a range of behavior, some of which can be predicted quantitatively or semi-quantitatively with molecular simulations, while others pose a challenge to capture at more than a qualitative level. Comparing a series of different coarse-grained models provides insight into the importance of domain structure, and balances between computational cost and the types of CG models that are used. In the case of single-chain antibodies, the results illustrate examples where the flexibility and dynamics of "linker" peptides can dominate the behavior, and pose a challenge for predicting the behavior and "developability" of candidate molecules.
Approximately 10-40% of the intra- and extracellular fluids of living organisms is occupied by macromolecules such as proteins, the internal dynamics of which is widely recognized as a crucial aspect for their function. The rather high concentration of such macromolecules is known as “macromolecular crowding” and was shown to influence reaction rates  and protein thermal stability. Here, we present a neutron backscattering study on the nanosecond self-diffusion of the antibody proteins immunoglobulins (Ig) in aqueous solution. We consider two systems: Ig and serum albumin (the two most abundant protein types in blood plasma), and Ig in cellular lysate, mimicking the cellular environment.
To investigate the effect of macromolecular crowding on protein dynamics in different environments, we systematically vary the concentration of Ig, serum albumin and cellular lysate, respectively. We find that, notwithstanding the different environments, the diffusion of Ig (as probed by neutron backscattering) as a function of the overall volume fraction is in rather good agreement with that of Ig in pure D2O as a function of its own volume fraction , pointing out the crucial role of hydrodynamics even in complex, biomimicking environments.
 Hall D. & Minton A. P. Biochim. Biophys. Acta 1649 (2003): 127.
 Grimaldo M., Roosen-Runge F., Zhang F., Seydel T., Schreiber F. JPCB 118 (2014): 7203.
Concentrated solutions of monoclonal antibodies have attracted considerable attention due to their importance in pharmaceutical formulations, yet their tendency to aggregate and the resulting high solution viscosity has posed considerable problems. It remains a very difficult task to understand and predict the solution behavior and stability of such solutions.
In this talk I will discuss a recent study  of the concentration dependence of the structural and dynamic properties of monoclonal antibodies using a combination of different scattering methods and microrheological experiments. The system is also investigated within a simple model of patchy colloids that incorporates the characteristic Y-shape of antibodies. To this aim we perform Monte Carlo simulations which are compared to analytical results, based on Wertheim theory applied to the case of hyperbranched polymers. Thanks to this colloid-inspired approach, we are able to disentangle self-assembly and intermolecular interactions and to describe the concentration dependence of structural and dynamic quantities such as the osmotic compressibility, the collective diffusion coefficient and the zero shear viscosity over the entire range of concentrations investigated.
Perspectives of this work will also be discussed.
 N. Skar-Gislinge, M. Ronti, T. Garting, C. Rischel, P. Schurtenberger, E. Zaccarelli and A. Stradner, to be submitted .
An outstanding challenge in computational biophysics is the simulation of a living cell at molecular detail. Over the past several years, using Stokesian Dynamics, progress has been made in simulating coarse grained molecular models of the cytoplasm. Since macromolecules comprise 20-40% of the volume of a cell, one would expect that steric interactions dominate macromolecular diffusion. However, the reduction in cellular diffusion rates relative to infinite dilution is due, roughly equally, to steric and hydrodynamic interactions, HI, with nonspecific attractive interactions likely playing rather a minor role. HI not only serve to slow down long time diffusion rates but also cause a considerable reduction in the magnitude of the short time diffusion coefficient relative to that at infinite dilution. More importantly, the long range contribution of the Rotne-Prager-Yamakawa, RPY, diffusion tensor results in temporal and spatial correlations that persist up to microseconds and for intermolecular distances on the order of protein radii. While HI slows down the bimolecular association rate in the early stages of lipid bilayer formation, they accelerate the rate of large scale assembly of lipid aggregates. This is suggestive of an important role for HI in the self-assembly kinetics of large macromolecular complexes such as tubulin. Since HI are important, questions as to whether continuum models of HI are adequate. Nevertheless, the stage is set for the molecular simulations of ever more complex subcellular processes and we discuss one such case, the diffusion of lac repressor in a packed cellular nucleoid.
There is a need for achieving high protein concentration liquid formulations of antibody therapeutics to meet patient dose requirements. Predicting the concentrated solution behaviour requires understanding how to map protein-protein interactions on simplified models, which account for the relative contributions from repulsive and attractive forces, shape and interaction anisotropy, and any effects due to intra-molecular flexibility. The overall aim here is to examine applicability and limitations of spherical versus anisotropic-shaped models for proteins in describing the thermodynamic properties and structure of concentrated solutions as probed by small angle X-ray scattering experiments on solutions of a monoclonal antibody or albumin. We show that an ellipsoidal versus spherical model provides an improved description for the excluded volume contribution to the thermodynamic properties and solution structure in concentrated albumin solutions. Molecular simulations of a three bead model for the antibody molecule, which is capable of reproducing generic features in the effective structure factor profile, indicates contributions from intra-molecular correlations can only be separated out for q values corresponding to characteristic separations greater than a protein diameter. Nevertheless fitting to integral equation calculations of the spherical structure factor over this limited q-range can still discriminate between steric-only models and models including an electrostatic repulsive potential with physically realistic charge parameters providing evidence that spherical models are accurate for interpreting longer-ranged repulsive forces.
We are conducting continuing studies of rotational diffusion, translational diffusion, and thermodynamic compressibility of the eye lens protein bovine gammaB crystallin at low and intermediate protein concentrations. For nuclear magnetic resonance (NMR) measurements, 15N-labeled bovine gammaB crystallin was produced in transformed E Coli by recombinant means, and isolated using size-exclusion and cation-exchange chromatography. For light scattering measurements, protein was isolated from young bovine lenses and isolated in the same fashion. Protein was concentrated for measurements in 10% D2O 100mM sodium phosphate buffer, pH 7.1, with 20mM dithiothreitol to inhibit oxidation. NMR transverse and longitudinal relaxation profiles were used to study concentration- and temperature-dependent rotational self-diffusion. Translational collective diffusion was measured with use of quasielastic light scattering, and solution Rayleigh ratios were measured using static light scattering. Characteristic rotational diffusion times of gammaB crystallin slowed from 9 nanoseconds to 13 nanoseconds, as protein concentration was increased from 0.25 to 2.6 millimolar. This slowing-down is well above effects due to solution viscosity increases in this concentration range. Evidence for concentration-dependent, non single-exponential rotational relaxation emerged and is awaiting follow-up experiments. Collective translational diffusion coefficient slowed from 9 x 10^(-11) m^2/s to 6 x 10^(-11) m^2/s over the same concentration range. The dependence of the Rayleigh ratio on concentration is consistent with attractive interactions, as characterized previously. We compare these results with computational hydrodynamic and theoretical calculations. To fully interpret the combined data, models for rotational diffusion in the presence of orientation-dependent direct as well as hydrodynamic inter-protein interactions may need further development.
The talk will present an introduction to x-ray photon correlation spectroscopy (XPCS) and the specific issues associated with XPCS measurements on biological macromolecules. This will include flux requirements and methods to ameliorate beam damage. XPCS measurements of the dynamics of concentrated suspension of eye-lens proteins will be presented. The measured time correlation functions from alpha crystallin suspensions will be compared with Langevin dynamics simulations. XPCS, dynamic light scattering and neutron spin echo measurements will be compiled to yield a comparison of concentrated alpha crystalline suspensions with hard sphere colloid theory over a wide range of length and time scales.
X-ray photon correlation spectroscopy (XPCS) measures nanoscale dynamics in real time by correlations of X-ray speckle patterns. The speckle patterns yield access to density-density correlation functions and also to higher order correlation functions. However, the highly intense X-ray beams of modern storage rings are also the cause for considerable radiation damage to the samples. Traditionally, XPCS experiments are performed with radiation doses of MGy to GGy, many orders of magnitude higher than tolerable for biological samples. We demonstrated recently  that XPCS experiments can be performed with very low doses reaching doses as low as a few kGy which opens the possibility to study dynamics of protein systems. In this talk I will present the methodology, opportunities, challenges and also first results of XPCS studies of radiation sensitive samples.
 J. Verwohlt et al. Phys Rev Lett 120, 168001 (2018)
Neutron scattering, when combined with computational science, can aid in the drug discovery process in several ways. We illustrate how precision neutron crystallography can permit the derivation of the thermodynamic driving forces behind the binding of drugs to their targets. Also, we show how computational drug design protocols benefit considerably from a dynamic, rather than just a static, description of the protein to be modulated. We describe how small-angle and dynamic neutron scattering, when combined with computer simulation, provide useful information on the motions involved. We show that motions in single protein molecules are complex, being non-ergodic and non-equilibrium, and exhibit ageing, these properties arising from the fractal nature of the topology and geometry of the energy landscape explored. We describe how taking these motions into account in supercomputer-based virtual high-throughput screening has led to the discovery of lead compounds for a variety of diseases.
Translational diffusion of macromolecules in cell is generally assumed to be anomalous due high macromolecular crowding of the milieu. Red blood cells are a special case of cells filled quasi exclusively (95 % of the dry weight of the cell) with an almost spherical protein: hemoglobin. Hemoglobin diffusion has since a long time been recognized as facilitating the rate of oxygen diffusion through a solution. We will address the question on how hemoglobin diffusion in the red blood cells can help the oxygen capture at the cell level and hence to improve oxygen transport. We have performed a measurement by neutron spin echo spectroscopy of the diffusion of hemoglobin in solutions with increasing protein concentration. We will show that hemoglobin diffusion in solution can be described as Brownian motion up to physiological concentration and that hemoglobin diffusion in the red blood cells and in solutions at similar concentration are the same. Finally, using a simple model and the concentration dependence of the diffusion of the protein reported here, we show that hemoglobin concentration observed in human red blood cells (≃ 330 g.L−1) corresponds to an optimum for oxygen transport for individuals under strong activity.
1 - S. Longeville and L. Stingaciu, Sci. Reports, 7 (2017) 10448
2 - A. Clark Jr., W. J. Federspiel, P. A. A. Clark and G. R. Cokelet, Biophys. J., 47 (1985) 171-181.
In a dense and crowded environment such as the cell, an individual protein feels the presence of surrounding proteins. It is thus expected that direct and hydrodynamic interactions strongly affect the diffusion of proteins. Examples are suspensions of eye lens proteins, where a dramatic slow down of the local short-time diffusion of γB-crystallin and a dynamical arrest is observed experimentally under crowded conditions. Here, we demonstrate that an application of colloid models, together with appropriate theoretical and simulation tools that allow to incorporate direct and hydrodynamic interactions, provides detailed insight into the dynamics of protein solutions. The hybrid simulation approach combines the multiparticle collision dynamics (MPC) method for the fluid with molecular dynamics simulations (MD) for the globular proteins. We present results for the short-time diffusion of different model proteins, where their dynamics are analyzed together with structural properties. The effect of shape anisotropy as well as weak attractive patches between colloids are discussed. In particular, we highlight the dramatic effect of weak interaction anisotropy known to exist between many globular proteins on the short-time diffusion under crowded conditions. This study is of great interest in applications such as formulations as well as for the fundamental understanding of soft matter in general and crowding effects in living cells in particular.
The function of the eye is dependent on a transparent, optically refractive, and deformable eye lens. These specific physical properties are realized by a crowded multicomponent mixture of mainly crystallin proteins within the cells in the eye lens. The underlying biophysical mechanisms are not only of fundamental interest, but highly relevant to better understand and treat eye conditions such as presbyopia and cataract.
We present experimental data on nanosecond dynamics in solutions of α, β and γ crystallins as model systems for the cytoplasm in the eye lens. While cage diffusion and gradient diffusion in α crystallin solutions are consistent with hard sphere systems [1,2], solutions of γ crystallins show clear signatures of short-range attraction, resulting in a significant slowing down of the cage diffusion compared to hard-sphere predictions , and critical slowing down of the gradient diffusion . β crystallins appear to have only weak attractive interactions, causing smaller deviations from hard sphere behavior than for γ crystallin.
Based on the dynamics characterised in mono-component solutions, we discuss the effects of mutual protein interaction in mixed solutions of crystallins on the dynamics and arrest behavior. For both mixtures of α / γ crystallins and β / γ crystallins, non-additive effects of the diffusion are observed, suggesting mutual interaction between the crystallins .
 S Bucciarelli, JS Myung et al. Sci. Adv. 2 (2016) e1601432
 G Foffi, G Savin, et al. PNAS 111 (2014) 16748
 S Bucciarelli, L Casal-Dujat, et al. JPCL 6 (2015) 4470
 A Stradner, G Foffi et al. PRL 99 (2007) 198103
Protein cluster formation plays an important role in some pathological pathways and in drug delivery applications. Studying protein diffusion allows to reliably monitor cluster formation, as exemplified in a study combining several scattering techniques to determine the cluster sizes in $\beta$-lactoglobulin . By investigating the short-time self-diffusion with incoherent quasi-elastic neutron backscattering (QENS) on the example of ovalbumin (OVA), we investigate crowding-induced cluster formation of OVA in aqueous (D$_2$O) solutions. To this end, the obtained diffusion coefficients are compared with the theoretical diffusion coefficients determined from the pdb structure of different cluster sizes. A monotonously increasing cluster size can be observed with increasing protein concentration within the accessed protein concentration range. While for low protein concentrations, the solution predominantly contains monomers, clusters larger than tetramers are observed for the highest investigated protein concentration. Different fit algorithms with and without imposing the dependence of the employed models on the scattering vector are applied, resulting in consistent results.
Besides the global diffusion providing information on the cluster size, information on the internal dynamics of the proteins is obtained simultaneously by the analysis of the QENS spectra. The internal dynamics of OVA is compared with the internal dynamics of bovine serum albumin, $\beta$-lactoglobulin and immunoglobulin.
 M. Braun et al., J. Phys. Chem. Lett. 8, 12, 2590-2596
Proteins have evolved into complex nanomachines able to couple dynamics over
many orders of magnitude in time to precisely and exquisitely control
chemical reactions and processes such as signal transfer and the generation
of mechanical force. To understand how they are able to achieve this
requires not only the determination of their structure, but also study of
their dynamics. Time-resolved structural biology is one route towards such
understanding, providing both high-resolution global structural information
and insight into dynamics. However, its application to a broad range of
proteins has been hampered by challenges in both reaction initiation and
access to high-brilliance sources with the needed photon flux for fast
time-resolved experiments. I will present our progress towards tackling both
these challenges by the development of new photocaging tools and the use of
multiplexing data collection strategies that enable fast time-resolved
experiments on weak photon sources with slow detectors.