Institute of Fundamental Technological Research

Membrane proteins, including G protein-coupled receptors (GPCRs), present a challenge in studying their structural properties under physiological conditions. Moreover, to better understand the activity of proteins requires examination of single molecule behaviors rather than ensemble averaged behaviors. Force–distance curve-based AFM (FD-AFM) was utilized to directly probe and localize the conformational states of a GPCR within the membrane at nanoscale resolution based on the mechanical properties of the receptor. FD-AFM was applied to rhodopsin, the light receptor and a prototypical GPCR, embedded in native rod outer segment disc membranes from photoreceptor cells of the retina in mice. Both FD-AFM and computational studies on coarse-grained models of rhodopsin revealed that the active state of the receptor has a higher Young's modulus compared to the inactive state of the receptor. Thus, the inactive and active states of rhodopsin could be differentiated based on the stiffness of the receptor. Differentiating the states based on the Young's modulus allowed for the mapping of the different states within the membrane. Quantifying the active states present in the membrane containing the constitutively active G90D rhodopsin mutant or apoprotein opsin revealed that most receptors adopt an active state. Traditionally, constitutive activity of GPCRs has been described in terms of two-state models where the receptor can achieve only a single active state. FD-AFM data are inconsistent with a two-state model but instead require models that incorporate multiple active states.

nanoindentation, rhodopsin, GPCR, membrane, biophysics, AFM, active, inactive, molecular dynamics, coarse graining, go-like model

We study the mechanical response of cellulose and β-amyloid microfibrils to three types of deformation: tensile, indentational, and shear. The cellulose microfibrils correspond to the allomorphs Iα or Iβ whereas the β-amyloid microfibrils correspond to the polymorphs of either two- or three-fold symmetry. This response can be characterized by three elastic moduli, namely, YL, YT, and S. We use a structure-based coarse-grained model to analyze the deformations in a unified manner. We find that each of the moduli is almost the same for the two allomorphs of cellulose but YL is about 20 times larger than YT (140 GPa vs. 7 GPa), indicating the existence of significant anisotropy. For cellulose we note that the anisotropy results from the involvement of covalent bonds in stretching. For β-amyloid, the sense of anisotropy is opposite to that of cellulose. In the three-fold symmetry case, YL is about half of YT (3 vs. 7) whereas for two-fold symmetry the anisotropy is much larger (1.6 vs. 21 GPa). The S modulus is derived to be 1.2 GPa for three-fold symmetry and one half of it for the other symmetry and 3.0 GPa for cellulose. The values of the moduli reflect deformations in the hydrogen-bond network. Unlike in our theoretical approach, no experiment can measure all three elastic moduli with the same apparatus. However, our theoretical results are consistent with various measured values: typical YL for cellulose Iβ ranges from 133 to 155 GPa, YT from 2 to 25 GPa, and S from 1.8 to 3.8 GPa. For β-amyloid, the experimental values of S and YT are about 0.3 GPa and 3.3 GPa respectively, while the value of YL has not been reported.

Tensile, shear, indentation, Atomic Force Microscopy, amyloid, cellulose

The application of coarse-grained (CG) models in biology is essential to access large length and time scales required for the description of many biological processes. The ELNEDIN protein model is based on the well-known MARTINI CG force-field and incorporates additionally harmonic bonds of a certain spring constant within a defined cutoff distance between pairs of residues, in order to preserve the native structure of the protein. In this case, the use of unbreakable harmonic bonds hinders the study of unfolding and folding processes. To overcome this barrier we have replaced the harmonic bonds with Lennard–Jones interactions based on the contact map of the native protein structure as is done in Go̅-like models. This model exhibits very good agreement with all-atom simulations and the ELNEDIN. Moreover, it can capture the structural motion linked to particular catalytic activity in the Man5B protein, in agreement with all-atom simulations. In addition, our model is based on the van der Waals radii, instead of a cutoff distance, which results in a smaller contact map. In conclusion, we anticipate that our model will provide further possibilities for studying biological systems based on the MARTINI CG force-field by using advanced-sampling methods, such as parallel tempering and metadynamics.

Martini force field, protein, molecular simulation, stretching AFM, large conformational changes

All-atom simulations are used to derive effective parameters for a coarse-grained description of the crystalline cellulose I

Cellulose, microfibril, allomorphs, structural transition, molecular dynamics, free energy

We construct two variants of coarse-grained models of three hexaoses: one based on the centers of mass of the monomers and the other associated with the C4 atoms. The latter is found to be better defined and more suitable for studying interactions with proteins described within α-C based models. We determine the corresponding effective stiffness constants through all-atom simulations and two statistical methods. One method is the Boltzmann inversion (BI) and the other, named energy-based (EB), involves direct monitoring of energies as a function of the variables that define the stiffness potentials. The two methods are generally consistent in their account of the stiffness. We find that the elastic constants differ between the hexaoses and are noticeably different from those determined for the crystalline cellulose Iβ. The nonbonded couplings through hydrogen bonds between different sugar molecules are modeled by the Lennard-Jones potentials and are found to be stronger than the hydrogen bonds in proteins. We observe that the EB method agrees with other theoretical and experimental determinations of the nonbonded parameters much better than BI. We then consider the hexaose-Man5B catalytic complexes and determine the contact energies between their the C4−α-C atoms. These interactions are found to be stronger than the proteinic hydrogen bonds: about four times as strong for cellohexaose and two times for mannohexaose. The fluctuational dynamics of the coarse-grained complexes are found to be compatible with previous all-atom studies by Bernardi et al.

Polysaccharide, protein, principal component analysis, coarse graining, molecular simulation

We propose to improve and simplify protein refinement procedures through consideration of which pairs of amino acid residues should form native contacts. We first consider 11 330 proteins from the CATH database to determine statistical distributions of contacts associated with a given type of amino acid. The distributions are set across the distances between the α-C atoms that are in contact. Based on this data, we determine typical radii of effective spheres that can be placed on the α-C atoms in order to reconstruct the distribution of the contact lengths. This is done by checking for overlaps with enlarged van der Waals spheres associated with heavy atoms on other amino acids.The resulting contacts can be used to identify non-native contacts that may arise during the time evolution of structure-based models. Here, the radii are used to guide reconstruction of nine missing side chains in a type I cohesin domain with the Protein Data Bank code 1AOH. We first identify the likely missing contacts and then sculpt the corresponding side chains by standard refinement tools to achieve consistency with the expected contact map. One ambiguity in refinement is resolved by determining all-atom conformational energies.

Cohesin, Go-like model, protein prediction, proteins, AFM, stretching

The analysis of the TEM images of indium tin oxide (ITO) clusters in ink solutions deposited from ink dispersions reveals that their geometry arises from a diffusion limited cluster aggregation (DLCA) process. We model films of ITO clusters as built through deposition of DLCA clusters made of primary spherical nanoparticles of 13 nm in diameter.
The deposition is then followed by a further compactification process that imitates sintering. We determine the conductivity of the sintered films by mapping the problem to that of the resistor network in which the contact regions between the touching spheres provide the dominant electric resistance. For a given volume fraction, conductivity of the sintered films is shown to be larger than that for the randomly packed spheres. However, the larger a typical radius of gyration of the clusters the smaller the enhancement. We also provide numerical tests for the routines used in the interpretation of the TEM images.

cluster aggregation, ITO