AbstractProtein disulfide isomerases (PDIs) family members are essential for proper folding of proteins entering the secretory pathway. Studies of these remarkable enzymes have shown that PDIs catalyze the oxidation, reduction and isomeration of disulfide bonds in the endoplasmic reticulum (ER). PDIs are found and conserved in a wide range of species and are ubiquitously expressed.In all eukaryotes, the ER redox potential must be tightly controlled since a large fraction of proteins within the cell require disulfide-bond formation. How PDIA1 maintains a balance between oxidation and reduction remains an important open question. Previous studies have revealed the tendency of PDIA1 to dimerize, but little is known about the functional relevance of PDIA1 dimerization. Also a recent crystal structure of a dimeric form of human PDIA1 shows that the formation of dimers inhibits substrate binding and therefore may function as a mechanism to regulate PDIA1 activity in the ER. This thesis has found that PDIA1 dimerizes in vivo and proposes that the dimerization of PDIA1 has physiological relevance by auto-regulating its activity. This mechanism would allow the ER to maintain a balance between oxidation and reduction necessary for native disulfide formation. Another enzyme, PDILT, is a testis-specific PDI. PDILT shares a high sequence similarity (32% identity) with PDIA1. Previous studies suggest that they share the same mechanism for binding substrates, which is mediated by a highly conserved hydrophobic pocket. In this thesis, the crystal structure of the PDILT b'-domain is reported and reveals a hydrophobic pocket that contains interesting features for substrate binding. Structural studies of this new PDI-member will help to understand the important role that PDILT plays in the differentiation and maturation on spermatozoids. The study of ER protein folding in the testis may lead to the identification of proteins and pathways associated with male infertility. / Résumé Les membres de la famille des protéines disulfures isomérases (PDI) sont essentiels pour le repliement des protéines entrant dans les voies de sécrétion. Les études sur ces enzymes remarquables ont montré que les PDI catalysent l'oxydation, la réduction et l'isomérisation des liens disulfures dans le réticulum endoplasmique (RE). Les PDI sont présentes et conservées chez une large gamme d'espèces et sont exprimées de manière omniprésente.Chez tous les eukaryotes, le potential rédox du RE doit être contrôlé fermement puisqu'une grande fraction des protéines dans la cellule nécessite la formation de liens disulfures. Le mécanisme avec lequel la PDIA1 maintient un équilibre entre l'oxydation et la réduction reste une question importante ouverte. Des études précédentes ont révélé que la PDIA1 a tendance à se dimériser, mais on sait peu de choses sur la pertinence fonctionnelle de la dimérisation de la PDIA1. De plus, une structure cristalline récente d'une forme dimérique de la PDIA1 humaine montre que la formation des dimères inhibe la liaison au substrat et ainsi peut agir comme un mécanisme de régulation de l'activité de la PDIA1 dans le RE. Cette thèse a trouvé que la PDIA1 se dimérise in vivo et propose que la dimérisation de la PDIA1 a une pertinence physiologique en auto-régulant son activité. Ce mécanisme permettrait au RE de maintenir un équilibre entre l'oxydation et la réduction nécessaires pour la formation de liaisons disulfures natives.Une autre enzyme, la PDILT, est spécifique pour le testicule. La PDILT partage une similitude de séquence importante (32%) avec la PDIA1. Des études précédentes suggèrent qu'elles partagent le même mécanisme de liaison des substrats qui est médiée par une poche hydrophobe hautement conservée. Dans cette thèse, la structure cristalline du domaine b' de la PDILT est reportée et révèle une poche hydrophobe qui contient des caractéristiques intéressantes pour la liaison du substrat. Des études structurelles de ce nouveau membre de PDI aideront à comprendre le rôle important que joue la PDILT dans la différentiation et la maturation des spermatozoïdes. L'étude du repliement des protéines du RE dans le testicule pourrait mener à l'identification de protéines et voies associées à l'infertilité masculine.

This thesis presents the application of k-space Image Correlation Spec- troscopy (kICS) to the analysis of fluorescence microscopy image time series for the measurement of particle diffusion in heterogeneous membranes, composed of micro- domains. The extension, testing and application of kICS for such measurements is developed both in silico with simulation and with in vivo cellular experiments.Connections between kICS analysis and other existing fluorescent microscopy techniques used in the study of heterogeneous membranes, such as single particle tracking (SPT) and spot vary Fluorescence Correlation Spectroscopy (FCS) are introduced. This is followed by the development of kICS theory of fluorescent particle diffusion within a heterogeneous two dimensional (2D) environment. Two possible membrane heterogeneities, isolated lipid micro-domains and actin proximal meshwork, are considered separately. The emergent models suggest that the kICS correlation function (CF) can be fit by a sum of two Gaussians in the case of particle diffusion in the presence of isolated micro-domains. These two fit components, called 'fast' and 'slow', with the fast associated with the rapid decay of the kICS CF at small spatial frequencies due to particle motion on large spatial scales outside domains while the slow component refers to the confined particle motion on large spatial frequencies or small spatial scales in domains. On the other hand, the meshwork confinement is well fit with a single Gaussian model for the analysis of kICS CF. These models suggest that the exponents and amplitudes of the fits embed the characteristic system parameters such as diffusion coefficients outside and inside domains, the partitioning rates, micro-domains radii and mesh pore size.Furthermore, systematic simulations to study different confinement scenarios were conducted and the calculated kICS correlation functions were fit and the output interpreted for recovery of self system parameters. The characterization of the simulated data suggests that kICS CFs exhibit various confinement dependent features, such as decays due to effective slow and fast dynamics populations and effective domain sizes. The in silico characterization of different confinement scenarios, suggests a connection between the apparent measured confinement properties, and the set system defining parameters. We explore the range and limits where confinement effects can be detected and accurately measured by kICS analysis. Possible systematic errors in the values of the fit extracted parameters due to background noise is discussed with possible alternative solutions.Finally, we apply this extension of kICS to the heterogeneous membrane en- vironment to explore the confinement dynamics of GPI-GFP anchored proteins in the basal plasma membrane of COS-7 cells. We employ a novel labelling approach of GPI-GFP using anti-GFP-Alexa594 and image the protein in COS-7 cell mem- branes with TIRF microscopy. Cells were exposed to enzymatic treatments, using the Cholesterol Oxidase (COase) and Sphingomyelinase (SMase), in order to dis- rupt membrane domains and change GPI-GFP confinement dynamics. We observe that GPI-GFP mobility and the effective domain size measured correlates with the enzymatic exposure time. We attribute it to the conversion of the membrane domain constituents, cholesterol and sphingomyelin, upon the enzymatic reactions, leading to membrane domain that are effectively larger and leakier. Finally, we conclude with possible improvements and future directions. / La thèse qui suit est a propos de l'adaptation de la technique de la spectroscopie par la corrélation des images dans l'espace de Fourier, appelle kICS. La nouveauté consiste en utilisation de kICS pour analyser les séries temporelles d'images fluorescentes afin de caractériser la diffusion des particules en présence des membranes hétérogénes, composées de micro-domaines.Tout d'abord, une parallèle est exposée entre l'analyse fondée sur kICS pro- posé ci-dessus et d'autres techniques de microscopie à fluorescence existantes et utilisées dans l'étude des membranes hétérogénes. Ensuite, on expose le développement de la théorie de kICS dans les cas de la diffusion des particules fluorescentes dans un espace hétérogène bidimensionnel (2D). Les deux hétérogénéités membranaires possibles, micro-domaines lipidiques isolés et le réseau de l'actine proximale, sont considérés séparément. Les modèles émergents suggèrent que la fonction de corrélation de kICS doit être caractérisé par une somme de deux Gaussiennes dans le cas de la dynamique des particules en présence de micro-domaines isolés. Ces deux éléments, appelés 'rapide' et 'lent', représentent les composantes dynamiques a deux échelles d'espace différentes. La rapide est associé à la décroissance rapide de la fonction de corrélation de kICS à petites fréquences spatiales dues au mouvement des particules sur de grandes échelles spatiales. La composante lente réfère au mouvement des particules confinées à des petites échelles spatiales, observées sur de grandes fréquences spatiales de kICS. D'autre part, la fonction de corrélation de kICS due au confinement par le réseau du cytosquelette peut être caractérise par unique décroissance Gaussienne. Ces modèles suggèrent que les exposants et les amplitudes obtenus par la caractérisation de la fonction kICS dépend des paramètres caractéristiques du système tels que les coefficients de diffusion à l'extérieur et à l'intérieur de domaines, les taux de migration de particules vers intérieur ou extérieur de micro-domaines ou des tailles de porosités du réseaux du cytosquelette.Les études systématiques par les simulations des scénarios différents de confinement et leurs effets sur la fonction de corrélation de kICS ont été explorés. La caractérisation des données simulées suggèrent que les fonctions de corrélation ont des caractéristiques qui dépendent de confinement et les propriétés spécifiques, tels que la dynamique des populations lents et rapides et la tailles effective de micro-domaines. La caractérisation des scénarios de confinement différents, représente les liens entre les propriétés apparentes mesurées de confinement, et un ensemble de paramètres définissant hétérogénéité. Nous explorons les limites pour lesquelles des effets de confinement ne sont pas observées dans la fonction de corrélation kICS. Les éventuelles erreurs systématiques dans les valeurs des paramètres extraits à cause du bruit de fond est discuté avec des possibles solutions. Finalement, nous utilisons l'analyse afin d'explorer la dynamique de confinement de la protéine ancrée à GPI-GFP dans la membrane plasmique basale des cellules COS-7. Nous explorons une approche nouvelle de la conjugaison entre le GPI-GFP et les anti-GFP-Alexa594 et imagé par la microscopie TIRF. Les cellules ont été exposées à des traitements enzymatiques, par Coase et SMase, afin de perturber domaines membranaires et changer la dynamique de confinement de GPI-GFP. Les réactions enzymatiques augmentent la mobilité et la taille effective des domaines de GPI-GFP. Nous attribuons cela à la conversion des constituants des domaines, le cholestérol et la sphingomyéline, par les réactions enzymatiques, ce qui conduit aux plus grandes et moins étanches domaines membranaires.

Small RNAs associate with Argonaute (AGO) proteins and guide them to their target nucleic acids via sequence specific interactions. Most small RNAs are formed from a double-stranded precursor. It is critical that the correct strand of this duplex is chosen and associates with an appropriate AGO protein. These processes are referred to as 'strand selection' and 'small RNA sorting' (or 'Argonaute sorting') and they are strongly influenced by the identity of the nucleotide found at the small RNA 5' end. The AGO MID domain, which is the site of interaction with the 5' nucleotide, was proposed to have evolved to preferentially bind certain 5' nucleotides and discriminate others. Therefore, the MID domains are believed to direct small RNA sorting and strand selection via specific interactions with the 5' nucleotide.I have solved the crystal structures of the MID domain from human AGO2 (hAGO2) in complex with nucleoside monophosphates (NMPs), which mimic the 5' nucleotide. The structures identified specific contacts between a rigid loop in the protein, which we termed the 'nucleotide specificity loop', and the nucleotide base of U or A, but not C or G. Dissociation constants from NMR titration experiments are in agreement with the crystal structures and this bias we observed in the hAGO2 MID domain reflects the distribution of nucleotides at the 5' end of miRNAs, which are the main binding partners for this AGO protein. I then characterized nucleotide-specific interactions in the MID domains of Arabidopsis thaliana (at)AGO1, atAGO2, and atAGO5. NMR titration experiments confirmed that these proteins display a bias towards U, A, and C, respectively. Furthermore, crystal structures revealed that the nucleotide specificity loop adopts strikingly different conformations in each of these MID domains. Structures of the atAGO1 MID domain in complex with NMPs revealed the structural basis for specificity of interaction with U or C and not A or G. A number of studies had implicated the MID domain of hAGO2 in miRNA-mediated translational repression via direct interactions with the mRNA 5' cap structure. Using cap-analogues I demonstrated that the interaction is non-specific and not physiologically relevant since it is located in the 5' nucleotide binding site. / Les petits ARNs s'associent aux protéines de la famille des Argonautes (AGO) et les guident vers les acides nucléiques ciblés par l'appariement de leurs séquences. La plupart des petits ARNs sont formés à partir d'ARN précurseurs à double brins. Il est critique que le bon brin de ce duplex soit choisi et qu'il s'associe à l'AGO approprié. Ces processus sont appelés 'sélection du brin' et 'triage du petit ARNs' (ou 'triage par l'Argonaute') et sont fortement influencés par l'identité du nucléotide se trouvant à l'extrémité 5' du petit ARN. Le domaine MID de l'AGO, étant le site de liaison avec le nucléotide 5', a été proposé d'avoir évolué afin d'interagir préférablement avec certains nucléotides 5' et de discriminer les autres. Ainsi, on croit que les domaines MID dirigent la sélection du brin et le triage du petit ARN en faisant appel à des interactions spécifiques avec le nucléotide 5'. J'ai résolu les structures cristallines du domaine MID de l'AGO2 humain (hAGO2) en complex avec des nucléosides monophosphates (NMPs), ces derniers imitant le nucléotide 5'. Les structures cristallines ont permis d'identifier des contacts spécifiques entre une boucle rigide de la protéine, appelée la 'boucle de spécificité au nucléotide', et la base des nucléotides U or A, mais pas C ou G. Les constantes de dissociation obtenues par des expériences de titration par RNM sont en accord avec les structures cristallines et la préférence observée chez le domaine MID d'hAGO2 reflète la distribution des nucléotides à l'extrémité 5' des micro (mi)ARNs, qui sont les principaux partenaires de liaison de cette protéine AGO. J'ai ensuite caractérisé les interactions spécifiques au nucléotide des domaines MID d'Arabidopsis thaliana (at)AGO1, atAGO2, et atAGO5. Des expériences de titration par RNM ont confirmé que ces protéines montrent une préférence pour U, A, et C, respectivement. De plus, les structures cristallines ont révélé que les boucles de spécificité au nucléotide adoptent des conformations considérablement différentes dans chacun de ces domaines MID. Les structures du domaine MID d'atAGO1 en complexe avec des NMPs ont révélé le fondement structural de la spécificité de l'interaction avec U ou C, et non A et G.Nombre d'études ont mêlé le domaine MID d'hAGO2 à la répression traductionnelle entremise par les miARNs par des interactions directes avec la coiffe à l'extrémité 5' de l'ARN messager. Par l'utilisation d'analogues de la coiffe, j'ai démontré que l'interaction est non-spécifique et n'est pas physiologiquement pertinente car elle est localisée dans le site de liaison avec le nucléotide 5'.

Streaming potentials arise from the presence of negative fixed charges in cartilage extracellular matrix glycosaminoglycans. Arthroscopic assessment of these potentials can potentially detect localized surface lesions and provide quantitative diagnostic information. Electrolyte filtration is also a phenomenon arising from glycosaminoglycans. Commonly assumed negligible despite a lack of experimental validation, it can be important for design and interpretation of streaming potential measurements and choice of modeling assumptions. The objective of this thesis was therefore to quantify electrolyte filtration and estimate its effect on streaming potential measurements. Chloride ion concentration in exudate of compressed cartilage was measured and explant GAG content was colorimetrically assayed. Pilot studies indicated that an appropriate strain rate for experiments was 8x10^(-3) s^(-1) in order to eliminate concerns of exudate evaporation and explant damage (at low and high strain rates, respectively). Exudate concentration of explants equilibrated in 1x PBS was significantly (p<0.05) lower than bath at strains of 37.5, 50 and 62.5%, with clear dependence on magnitude. Exudate concentration was also significantly lower than that of the bath when 50% strain was applied after equilibration in 0.5, 1 and 2x PBS while the relative difference seemed to increase with decreasing bath concentration (p=0.065 between 0.5 and 2x PBS). Decreasing exudate concentration correlated negatively with increasing post-compression GAG concentration, while no difference between exudate concentration and bath concentration was ever observed for compression of uncharged agarose gel controls. Findings show that exudate from compressed cartilage is dilute relative to bath due to the presence of matrix fixed charges. This difference leads to the generation of extraneous diffusion potentials during streaming potential measurements, particularly under conditions of low strain rates and high strains. / Le potentiel d'écoulement est due à la présence de charges fixes négatives sur les glycosaminoglycanes de la matrice extracellulaire du cartilage. Son évaluation arthroscopique peut potentiellement détecter les lésions localisées et fournir de l'information diagnostique quantitatif. La filtration d'électrolyte est également un phénomène découlant des glycosaminoglycanes. Généralement supposé négligeable malgré un manque de validation expérimentale, il peut être important pour la conception et l'interprétation des measurements du potentiel d'écoulement et pour le choix des suppositions de modélisation. L'objectif de cette thèse est donc de quantifier le phénomène de filtration d'électrolyte et d'estimer son effet sur les mesures de potentiel d'écoulement. La concentration de chlorure dans l'exsudat de cartilage comprimé a été mesurée et le contenu de GAG a été colorimétriquement dosé. Des études pilotes ont indiqué qu'une vitesse de déformation de 8x10^(-3) s^(-1) était appropriée afin d'éliminer les préoccupations d'évaporation d'exsudat ainsi que du dommage à l'explant (à haute et basse vitesses, respectivement). La concentration d'exsudat des explants équilibrés dans 1x PBS et soumis à une deformation de 37,5, 50 et 62,5% était significativement (p<0.05) inférieure à celle du bain, avec claire dépendance sur la magnitude de déformation. Elle était également significativement inférieure lorsqu'une deformation de 50% a été appliquée après l'équilibrage en 0,5, 1 et 2x PBS. La différence relative avec le bain semblait augmenter avec la diminution de la concentration de celui-ci (p=0,065 entre 0,5 et 2 PBS). La baisse de concentration de l'exsudat a corrélé négativement avec le contenu de glycosaminoglycanes post-compression, alors qu'aucune différence n'a jamais été détectée dans l'exsudat des explants d'agarose. Nos résultats démontrent que la concentration d'exsudat de cartilage comprimé est dilué par rapport au bain en raison de la présence des charges fixes. Cette différence mène à la génération de potentiels de diffusion durant la mesure du potentiel d'écoulement, surtout sous les conditions de haute déformation et de faible vitesse de deformation.

While there has been much speculation on the role the lipid matrix plays in controlling and mediating the inter-molecular dynamics of proteins, there has been little direct evidence for how the lipid bilayer can mediate the supra-molecular organization of these proteins. With the technique of X-ray in-plane scattering, we can directly measure the correlation function of gramicidin channels in the plane of the bilayer. We have studied the simple case of gramicidin incorporated into model Dilauroyl and Dimyristoyl-Phosphocholine (DL-,DMPC) membranes, and have seen how changing the membrane thickness changes peptide-peptide correlation. In the fluid phase, DLPC and DMPC present an elastic medium that can perhaps match the hydrophobic length of the peptide. The resulting membrane deformation energy is expected to cause inter-peptide attraction. Indeed, the scattering data shows that the gramicidin-gramicidin correlation distance is smaller in DMPC than in DLPC. Thus we have obtained direct evidence of membrane mediated peptide-peptide interaction. As the temperature drops and DMPC approaches its main transition temperature, gramicidin exhibits close packing. This may indicate that the channels are being excluded from gel lipid regions and are packing closer together.

<p> Predicting protein-ligand binding free energy from physical principles is a grand challenge in biophysics, with particular importance for drug discovery. Free energy calculations compute binding affinities by using classical mechanics to model the protein and ligand at atomic resolution, and using statistical mechanics to analyze simulations of these models. The binding affinities computed from these simulations are fully rigorous and thermodynamically correct <i>for the model</i> (with adequate sampling), and will agree with experimentally measured binding affinities if the model is accurate. Because free energy calculations capture the full statistical complexity of binding for flexible molecules at ambient temperature, they offer the greatest potential for quantitative accuracy of any physical method for predicting binding. </p><p> Here, I (&amp; coauthors) present several studies relating to using free energy calculations to predict protein-ligand binding affinities for charged compounds. First, we introduce the Separated Topologies method, an approach for using free energy calculations to predict relative binding affinities of unrelated ligands. This method is useful for studying charged compounds because charged compounds are very difficult to study using absolute binding calculations, increasing the importance of relative binding calculations. Second, we use free energy calculations to predict absolute binding affinities for charged molecules to a simplified protein binding site, which is specially designed for studying charged interactions. These predictions are compared to new experimental affinity measurements and new high-resolution structures of the protein-ligand complexes. We find that all affinities are predicted to be too strong, and that this error is directly correlated with the polarity of each ligand. By uniformly weakening the strength of electrostatic interactions, we are more successful at predicting binding affinity. Third, we design and validate an analytical correction scheme to correct binding free energy calculations of ions for artifacts caused by the periodic boundary conditions employed in simulations. Fourth, we examine the sensitivity of binding affinities from free energy calculations to the force field parameters used in the simulations. This provides insight into the strength of electrostatic interactions in protein simulations, complementing our previous work comparing simulation results to experiments. Finally, we discuss potential future directions of this work. </p>

We investigated the structural basis of calmodulin (CaM)-mediated regulation of the skeletal muscle ryanodine receptor (RyR1), a calcium channel that plays a key role in excitation-contraction coupling in muscle cells. In order to understand the complex interaction of CaM with this receptor, we pursued NMR and X-ray crystallographic studies of CaM/RyR1 peptide complexes both in the presence and in the absence of calcium. We have determined the 2.0 A crystal structure of Ca2+CaM in complex with a 30-residue peptide corresponding to the binding region for CaM on RyR1 (residues 3614 to 3643). The structure reveals that hydrophobic anchor residues in the target arranged in a novel '1-17' spacing allow each calmodulin lobe to interact with the peptide independently. Solution NMR 15N relaxation measurements and residual dipolar couplings confirm the structure of each calmodulin lobe and show that the complex undergoes segmental domain motion. Fluorescence measurements indicate that CaM binds with both domains to the 3614-3643 peptide, whereas if the second anchor is unavailable, CaM can bind without collapsing on the target. The independence of the two lobes of calmodulin offers a structural explanation for how other domains may compete for binding to this region to regulate the channel. NMR studies of these interaction partners indicate that the RyR1 target binds to only the C-lobe of CaM at low calcium concentrations, similar to other ion channels whose activity is modulated by CaM. The conformation of the C-lobe in the calcium-free complex closely resembles the one seen in the crystal structure of the calcium-loaded complex, suggesting that binding of the RyR1 peptide locks the C-domain of CaM in a conformation similar to that of the calcium-loaded protein. Comparison of the CaM/peptide complexes at low and high calcium concentrations provides a model for how CaM interacts with this region of RyR1: the C-lobe is constitutively tethered to the 3614-3643 target and is calcium-loaded even at low, resting calcium levels, whereas high calcium induces the N-lobe to bind to this region. In this way, the N-lobe of CaM acts as a Ca2+ sensor for RyR1 by switching between different binding sites on the receptor.

The detailed characterization of the overall folding landscape of a biologically relevant protein is an outstanding challenge to both all-atom simulation and experiment. No single technique, either experiment or simulation, at present can explore the wide range of time and length scales spanned by a protein molecule during its folding. This limitation advocates for the development of novel methodologies that can combine the complementary strengths of experiment and simulation for a complete characterization of protein landscapes at multiple time- and length- scales. This thesis focuses on providing a realistic, but simplified, description of protein landscapes that can be used as a solid starting point toward the development of such a multi-resolution framework. Toward that goal, we have successfully characterized the complex folding landscape of a monomeric lactose repressor protein by combining experimental data with simulation results, as obtained using a structure-based simplified protein model. In addition, we have developed a realistic, but coarse-grained, protein model that contains the crucial physical-chemical ingredients shaping protein landscapes. The simulated folding landscapes of a number of proteins obtained using this simplified model show remarkable quantitative agreement with experimental measurements. This minimalist model has further allowed investigation of the direct connection between folding and signaling of a photoreceptor protein, providing valuable insight into the protein folding-function relationship. The choice of a few optimal reaction coordinates is crucial for identifying the critical transition regions on the folding landscape of a protein. We have developed a powerful technique to automatically extract a set of collective coordinates from the configurational sampling generated during a molecular simulation. The results show that the coordinates emerging from this technique can accurately describe a complex protein folding reaction. Finally, an efficient multi-scale simulation procedure is proposed that can precisely identify the folding transition regions of a protein. In practice, the averaged coarse-grained description of the fast protein dynamics, as extracted from an ensemble of short MD trajectories, is used to globally trace the underlying effective folding free energy landscape.

The voltage-dependent movement, or electromotility, of cochlear outer hair cells (OHCs) contributes to cochlear amplification in mammalian hearing. Prestin, a transmembrane protein expressed in the lateral wall of the OHSs, is essential for electromotility, but molecular details of its function are unknown. Using a combined optical tweezers and whole-cell voltage clamping system, we have investigated the contribution of prestin to the mechanics and electromechanical force (EMF) of membrane tethers, and related these parameters to prestin-associated non-linear capacitance (NLC). Our study included three test groups consisting of (i) untransfected; (ii) wild type (WT) prestin-transfected; and (iii) mutant (A100W) prestin-transfected human embryonic kidney (HEK) cells. While there was no difference in membrane effective viscosity among the three cell types, tethers from WT prestin-transfected HEK cells demonstrated altered membrane mechanical parameters and increased EMF values than control untransfected HEK cells. We found that A100W mutation in prestin, which eliminates NLC also diminishes EMF, but does not affect membrane mechanical parameters. These results suggest that prestin-associated charge transfer is associated with generation of EMF in the membrane, but independent of the effect of prestin on membrane mechanics. Based on our results, we propose synergistic effects of prestin and the membrane in the generation of NLC and electromotility. The information these results provide is important for understanding protein/membrane interactions, prestin properties and the origin of electromotility.

Multi-particle non-equilibrium dynamics in two-channel biological transport systems is investigated theoretically within the framework of asymmetric simple exclusion processes (ASEP). In exclusion processes particles move along the lattice by hopping between neighboring sites that are vacant. We consider the systems with open boundaries, where particles enter the lattice on the entrance site and leave from the exit site with given rates. Four different ASEP models are studied. The first two models investigate interchannel coupling between parallel channels in a one-way transport system. The third model considers junction of two parallel tracks, while the last model investigates two-way transport system with narrow entrances with coupling on the boundaries. Theoretical investigation of these non-equilibrium systems reveal many interesting phenomena, such as unusual phase diagrams that contain up to seven stationary-state phases, localization of the domain wall in the bulk of the system, symmetry-breaking and strong interparticle correlation. Stationary phase diagrams, particle currents and bulk values of densities are calculated in a mean-field approximation for the systems in the thermodynamic limit. In addition, exact matrix product ansatz method and phenomenological domain-wall theory are applied to analyze dynamic properties. For several systems nearest-neighbour correlation and density distribution functions are computed and size-scaling effects are analyzed. Extensive Monte Carlo computer simulations are carried out for all systems to test predictions and they verify our theoretical results.