Session V18: Biopolymer Molecules - Solutions, Networks, and Gels

The Effects of Filament Aging and Annealing on a Lamellipodium Undergoing Disassembly By Severing

We construct a simplified model of a lamellipodium and use a numerical simulation to study its properties as it disassembles by filament severing. The growing lamellipodium is modeled as a 2D or 3D periodic lattice of crosslinked actin filaments. At each time step a new layer of actin filaments is added at the membrane, existing filaments are severed stochastically, and disconnected sections of the network are removed. Filament aging is modeled by including several different filament chemical states. Filament annealing is included by allowing existing filaments to grow new filaments. The properties of the model are studied as functions of the number of states and the severing and annealing rates. We find that the network width is proportional to the sum of the average lifetimes of the states, and is well modeled by a simple kinetic theory. The edge of the growing network becomes sharper as either the number of states or the dimensionality is increased. Annealing increases the average length of the network, and we find that the network length diverges at a critical annealing rate.   

Studying the formation of different phases of self-assembled cross-linked F-actin

Self-assembly of the system of F-actin and different linking proteins is studied using complementary methods of confocal microscopy, small angle x-ray scattering (SAXS) and molecular dynamics (MD) simulations. Studies using alpha-actinin (as a cross-linker) show that, by varying the actin concentration (C$_{A})$ and $\alpha $-actinin to actin molar ratio ($\gamma )$ the assembled system might fall in one of three different phases: (1) loosely connected network of F-actin and bundles, (2) strongly connected and homogeneous network of bundles, and interestingly, (3) loosely connected and inhomogeneous network of dense domains -- an intermediate phase between the first two. The phenomena can be explained statistical mechanically and replicated using our MD simulations. Further understanding based on simulations with different types of cross-linkers shows that the formation of different phases is related to the flexibility in binding between F-actin and cross-linkers, which leads to the possibility of forming branching points and thus bundle networks.   

A structural study of F-actin - filamin networks

The cell's ability to move and contract is attributed to the semi-flexible filamentous protein, F -actin, one of the three filaments in the cytoskeleton. Actin bundling can be formed by a cross-linking actin binding protein (ABP) filamin. By examining filamin's cross-linking abilities at different concentrations and molar ratios, we can study the flexibility, structure and multiple network formations created when cross-linking F-actin with this protein. We have studied the phase diagram of this protein system using fluorescence microscopy, analyzing the network structures observed in the context of a coarse grained molecular dynamics simulation carried out by our group.   

The mobility of individual filaments in microtubule and actin bundles

Friction between biopolymers is not well understood. We investigated the mobility of individual filaments in actin and microtubule based bundles. We find that microtubules, bundled under the influence of low concentrations of depleting agent, slide with respect to each other due to thermal effects. Using darkfield microscopy and filament-tracking software we analyzed this sliding as 1-dimensional diffusion. Our analysis reveals that this process is subdiffusive, likely due to a stick-slip mechanism. In contrast, actin filaments within a bundle do not exhibit any sliding motion. We therefore used laser tweezers to actively pull actin filaments past each other at a constant velocity. The resulting data showed that force scales linearly with velocity and overlap length of the filaments.   

Characterization of Hyaluronan-Protein Microstructures and Polymer Solutions

Evidence is mounting that mechanical and topographical features of biomaterials can be as critical for cellular behavior as chemical properties. A case in point is hyaluronan (HA), a large polysaccharide with unique mechanical and hydrodynamic properties, found in many tissues and bodily fluids. Thanks to a large variety of accessible conformations and aggregation states, this remarkable polymer can impart on its biological environment a diverse range of structural and viscoelastic properties with far-reaching consequences for cell physiology (migration, inflammation, cancer). Supramolecular assembly of HA is typically mediated by HA-binding proteins. These specialized molecules are known to assist the formation of organized structures, such as cross-linked bundles, gels, or the all-important pericellular coat, a polymer network anchored to many cell surfaces. Precisely how the material properties of HA-rich matrices and aggregates are modified by the associated proteins, however, is largely a matter of speculation. We will present new insights concerning the cell coat and HA-protein solutions characterized using passive microrheology, fluorescence recovery after photobleaching (FRAP), and optical force probe microscopy.   

Unbinding of Semi-flexible Bio-polymers from Columnar Traps: An Exactly Solvable Problem of Statistical Mechanics

We elucidate unbinding of long semi-flexible bio-polymers from long line-like attractive potential wells (columnar traps). This phase transition has been observed in the experiments with DNA molecules adsorbed on micro-structured supported cationic lipid membranes. It provides a new way to stretch (linearize) single DNA coils [Hochrein, Leierseder, Golubovic, and Raedler, Physical Review E (2007)]. We reveal that this phase transition is an exactly solvable problem of statistical mechanics. Our theory is based on mapping this nontrivial problem onto a novel class of directed random walks. This DNA unbinding transition turns out to have a unique thermodynamic character: It is of the second order however with very weakly (logarithmically) diverging correlation length. This feature induces a very strong divergence of the heat capacity at the unbinding transition of a semi-flexible polymer from a columnar trap. Our exact solution of this statistical physics problem opens a new venue in the theory of molecular shape control of bio-polymers such as DNA molecules adsorbed on specially designed biocompatible surfaces.   

Semi-flexible Polymer Dynamics in Nanoslits Smaller than the Persistence Length

Theories have been developed for the dynamics of confined semi-flexible polymers. When the confinement length is smaller than the chain radius of gyration (Rg) and much greater than the Kuhn length (bk), the blob scaling regime characterizes the chain dynamics, while Odijk's theory characterizes chain relaxation when the confinement length is much smaller than bk. Nevertheless, the boundary of these two regimes is not well-defined. This work will characterizes the transition from the blob regime dynamics to the Odijk regime dynamics using Brownian dynamics simulations coupled with the lattice Boltzmann method. In addition, we will examine the effects of confinement on the screening of intra-chain segment hydrodynamic interactions and the distribution of segments in sub-persistence length confinement.   

Gap Dependent Rheology in {\em Type I} Collagen Gels

Branched {\em type I} collagen fiber networks provide extracellular support in mammalian tissues. The intricate network structure can succumb to partial or complete tearing under sufficient applied strain. Under small shear strains, {\em in vitro} collagen gels exhibit strain-stiffening while maintaining overall network integrity. Higher shear strains lead to network failure through discrete yielding events. We perform rheology and confocal-rheology experiments to fully elucidate the strain-stiffening and yielding behavior in these highly nonlinear materials. We apply continuous shear strains to collagen gels confined within the rheometer at fixed gaps. We observe that sheared collagen in the strain-stiffening and yielding regime has an apparent modulus that is strongly dependent on the collagen thickness. Moreover, we demonstrate that network yielding is universally controlled by the ratio of the collagen thickness to the mesh size. These results have broad implications for the interpretation of rheological data of extracellular matrix proteins and for the design of biomimetic scaffolds.   

Fundamental studies on shear-thinning and self-healing peptide hydrogels

Peptides have been designed to fold into beta-hairpins once exposed to physiological conditions, and then subsequently self-assemble into stiff hydrogels stabilized by physical cross-links. These physical gels shear-thin and flow like a solution under a proper shear stress. However, as soon as the stress is removed, the shear-thinned gel solutions immediately self-heal into solid gels with gel rigidity restoring over time which suggests the possibility of delivering the hydrogel construct with a desired therapeutic payload encapsulated toward an in vivo site by syringe injection. Current rheometric results indicate that gel restoration kinetics is the same whether the shear stress is applied by the rheometer or via syringe injection. Under steady shear flow, the gel network morphology was studied by flow-SANS and the motion of gel was visually monitored by rheo-microscopy. Confocal microscopy was used to track the flow of the hydrogels through a channel. The results explain how the gel network morphology evolves during shear-thinning and subsequent rehealing process.   

Structure-Properties Relationship of Hierarchically Ordered Large and Small Molecules Self-Assemblies.

Membranes formed by mixing high molecular weight hyluronic acid (HA) and oppositely charged peptide amphiphiles (PAs) have been shown to have a unique hierarchically ordered structure which consists of three regions: an amorphous biopolymer layer, a narrow region of PA fibers parallel to the interface and a layer of fibers perpendicular to the interface. Understanding of the structure-property relationships in these self-assembling systems is a necessary step in designing these structures for specific applications. We have formed and characterized PA/polymer self-assembled membranes using different polyelectrolytes (alginate, $\lambda $-carrageenan, poly(acrylic acid) etc). SEM micrographs show that these assemblies have the same parallel/perpendicular fibers structure as the original HA/PA assembly. The mechanical properties and water permeability of these structures measured by membrane inflation techniques and osmotic swelling indicate that the polymer characteristics [i.e. Mw, charge density] are an important factor in determining structure formation, kinetics and final properties.   

Protein Structure and Stability in Neat Ionic Liquid

Ionic liquid (IL) as a medium for room temperature preservation of biomacromolecules has been proposed, and to investigate the possibility, we studied physicochemical and enzymatic properties of several proteins in the neat hydrophilic IL, ethylmethyl imidazolium ethyl sulfate [EMIM][EtSO4]. Molecular dissolution of $\alpha $-chymotypsin, cytochrome-c and other proteins could be achieved with moderate heating (60C). Dynamic light scattering and dilute solution viscometry typically reveal protein size slightly larger than in buffer, suggesting different solvation or protein unfolding. Spectroscopic methods (UV-Vis, fluorescence, FTIR, CD) show largely unchanged secondary structure but significantly changed tertiary structure. IL-dissolved cytochrome-c has heightened peroxidase activity, supporting the same conclusions. Transfer of dissolved protein from IL to buffer and ensuing alterations to protein conformation/activity will be discussed.   

Biomimetic surface coatings from modular amphiphilic proteins

Recombinant DNA methods have been used to develop a library of diblock protein polymers for creating designer biofunctional interfaces. These proteins are composed of a surface-active, amphiphilic block joined to a disordered, water soluble block with an end terminal bioactive domain. The amphiphilic block has a strong affinity for many synthetic polymer surfaces, providing a facile means of imparting biological functionality to otherwise bio-neutral materials through physical self-assembly. We have incorporated a series of bioactive end domains into this diblock motif, including sequences that encode specific cell binding and signaling functions of extracellular matrix constituents (e.g. RGD and YIGSR). In this talk, we show that these diblock constructs self-assemble into biofunctional surface coatings on several model synthetic polymer materials. We demonstrate that surface adsorption of the proteins has minimal impacts on the presentation of the bioactive domains in the soluble block, and through the use of microscopic and cell proliferation assays, we show that the resulting biofunctional interfaces are capable of inducing appropriate cellular responses in a variety of human cell types.   

Sensitive and selective protein detection by a molecular imprint nanosensor

It has been more than thirty years since the first report of the molecular imprint technique. However the progress towards imprint-based protein sensing has been fairly slow. Here, we report a significant advance on sensitivity and selectivity of imprint technique for protein detection: 1e4 times more sensitive than the state of the art. The lowest detectable human ferritin concentration was 10 pg/L by measuring the capacitance and resistance change in the sensor. Robust selectivity was demonstrated using other proteins, binary samples, and cellular protein extracts. Evaluation with calmodulin revealed that protein conformational changes could be detected. The molecular imprinting component of the nanosensor affords detection of a range of molecules, including macromolecules in a label-free manner and should prove useful in those instances where antibodies, aptamer, or natural ligands are not available.   

Imaging the Enzymatic Degradation of Individual Bundles of Cellulose Fibers

An automated angle scanning surface plasmon resonance (SPR) imaging experiment was designed and constructed. The experiment enables the accurate tracking of multiple regions of interest (ROIs) on the sample as a function of time. This allows the accurate tracking with time of changes to surfaces that are inherently laterally inhomogeneous. We have used the SPR imaging experiment to study the interaction of enzymes with bundles of cellulose fibers that have been heterogeneously distributed on a gold film coated with a thin layer of thioglucose. These data have allowed us to monitor the adsorption of the enzymes and subsequent degradation of the cellulose fibers in real time on samples that are of relevance to the cellulosic ethanol industry.   

Convection driven generation of long-range material gradients in microchannels

Natural materials exhibit anisotropy with variations in soluble factors, cell distribution, and matrix properties. The ability to recreate the heterogeneity of the natural materials is a major challenge for investigating cell-material interactions and for developing biomimetic materials. Here we present a generic fluidic approach using convection and alternating flow to rapidly generate multi-centimeter gradients of biomolecules, polymers, beads and cells and cross-gradients of two species in a microchannel. Accompanying theoretical estimates and simulations of gradient growth provide design criteria over a range of material properties. A poly(ethylene-glycol) hydrogel gradient, a porous collagen gradient and a composite material with a hyaluronic acid/gelatin cross-gradient were generated with continuous variations in material properties and in their ability to regulate cellular response. This simple yet generic fluidic platform should prove useful for creating anisotropic biomimetic materials and high-throughput platforms for investigating cell-microenvironment interaction.