Session P34: Biopolymers and Biohybrid Polymers - Assembly and Thermodynamics

Thermodynamics, morphology, and kinetics of early- stage self-assembly of pi-conjugated oligopeptides

Synthetic oligopeptides containing $\pi$-conjugated cores self-assemble novel materials with attractive electronic and photophysical properties. All-atom, explicit solvent molecular dynamics simulations of Asp-Phe-Ala-Gly-OPV3-Gly-Ala-Phe-Asp peptides were used to parameterize an implicit solvent model to simulate self-assembly. At low-pH conditions, peptides assemble into $\beta$-sheet-like stacks with strongly favorable monomer association free energies of $\Delta F \approx -25 k_B T$. Aggregation at high-pH produces disordered aggregates destabilized by Coulombic repulsion between negatively charged Asp termini. We model simulations of hundereds of monomers as a continuous-time Markov process. We infer transition rates between different aggregate sizes and microsecond relaxation times for early-stage assembly. Our data suggests a hierarchical model of assembly in which peptides coalesce into small clusters over tens of nanoseconds followed by structural ripening and diffusion limited aggregation on longer time scales. This work provides new molecular-level understanding of early-stage assembly, and a means to study the impact of peptide chemistry upon the thermodynamics, assembly kinetics, and morphology of the supramolecular aggregates.   

Bridging Length Scales to Study Self-Assembly and Self-Organization

A variety of proteins can assemble into large polymers as an integral part of their biological function. Studying the biochemistry and biophysics of polymer formation often involves time-resolvable measurements of the amount of polymer. Non-invasive measurements of polymer can be divided into two categories: short (spectroscopy) and large (microscopy) length scale measurements. Microscopy-based estimates of polymer amount are often dependent on spatial non-uniformity of polymer, whereas spectroscopy-based estimates of polymer amount are often based on models that are difficult to test. Here we show how both large and small length scale measurements can be combined to validate the assumptions behind both measurements while incorporating both measurements to make more accurate estimates of polymer amount. We utilize this approach with two-photon microscopy and FRET to measure the amount of tubulin (monomer) in microtubules (polymer) in order to study microtubule nucleation in cell extracts. In addition, this approach may be useful to study a wide variety of polymers, including actin filaments, viruses, lipid membranes, and other protein aggregates.   

Creating Ordered Antibody Arrays with Antibody-Polymer Conjugates.

Antibodies are a category of functional proteins that play crucial roles in the immune system and have been widely applied in the area of cancer therapeutics, targeting delivery, signal detection, and sensors. Due to the extremely large size and lack of specific functional groups on the surface, it is challenging to functionalize antibodies and manipulate the ordered packing of antibodies in an array with high density and proper orientation, which is critical to achieve outstanding performance in materials. In this work, we demonstrate an efficient and facile approach for preparing antibody-polymer conjugates with two-step sequential ``click'' reaction to form antibody-polymer block copolymers. Highly ordered nanostructures are fabricated based on the principles of block copolymer self-assembly. The nanostructures are studied with both small angle X-ray scattering (SAXS) and transmission electron microscopy (TEM). Lamellae with alternating antibody domain and polymer domain are observed with an overall domain size of \textasciitilde 50 nm. The nanostructure not only increases the packing density and promotes proper orientation of the antibody, but also provides possible channel to facilitate substrate transportation and improves the stability of the antibody.   

Self-assembly of Artificial Actin Filaments

Actin Filaments are long, double-helical biopolymers that make up the cytoskeleton along with microtubules and intermediate filaments. In order to further understand the self-assembly process of these biopolymers, a model to recreate actin filament geometry was developed. A monomer in the shape of a bent rod with vertical and lateral binding sites was designed to assemble into single or double helices. With Molecular Dynamics simulations, a variety of phases were observed to form by varying the strength of the binding sites. Ignoring lateral binding sites, we have found a narrow range of binding strengths that lead to long single helices via various growth pathways. When lateral binding strength is introduced, double helices begin to form. These double helices self-assemble into substantially more stable structures than their single helix counterparts. We have found double helices to form long filaments at about half the vertical binding strength of single helices. Surprisingly, we have found that triple helices occasionally form, indicating the importance of structural regulation in the self-assembly of biopolymers.   

Intermediate-filaments: from disordered building blocks to well-ordered cells.

In the past decade it was found that $\approx $50{\%} of human proteins contain long disordered regions, which play significant functional roles. As these regions lack a defined 3D folded structure, their ensemble conformations can be studied using polymer physics statistical-mechanics arguments. We measure the structure and mechanical response of hydrogels composed of neuronal intermediate filaments proteins. In the nervous system, these proteins provide cells with their mechanical support and shape, via interactions of their long, highly charged and disordered protein chains. We employ synchrotron small-angle X-ray scattering and various microscopy techniques to investigate such hydrogels from the nano- to the macro-scale. In contrast to previous polymer physics theories and experiments, we find that shorter and less charged chains can promote network expansion. The results are explained by intricate interactions between specific domains on the interacting chains, and also suggest a novel structural justification for the changing protein compositions observed during neuronal development. We address the following questions: Can protein disorder have an important role in cellular architecture? Can structural disorder in the micro-scale induce orientational and translational order on the macro-scale? How do the physical properties of disordered protein regions, such as charge, length, and hydrophobicity, modulate the cellular super-structure?   

The effect of local melting of DNA on DNA loop formation

Statistical mechanics of double-stranded DNA (dsDNA) is well described by the wormlike chain model (WLC) which assumes a harmonic bending potential. Such smooth bending potential may no longer be valid for large bending angles to form small loops (\textless 100 bp). Instead, DNA may rely on rare structural transitions such as local melting (opening) of base pairs to lower the energetic cost. In theory, open base pairs called bubbles can increase the looping probability of short DNA molecules by a few orders of magnitude, but a robust experimental validation of this theoretical prediction is lacking. Here, we investigated the correlation between local melting probability and looping dynamics of dsDNA using single-molecule fluorescence resonance energy transfer (FRET). We designed two groups of short DNA molecules with low and high melting probabilities around their center and measured their looping and unlooping rates in equilibrium. Our data allow rigorous tests of meltable wormlike chain (MWLC) models at short length scales for setting ranges of acceptable free energy cost of bubble formation and flexibility values of a bubble.   

Effect of Backbone Design on Hybridization Thermodynamics of Oligo-nucleic Acids: A Coarse-Grained Molecular Dynamics Simulation Study

DNA hybridization is the basis of various bio-nano technologies, such as DNA origami and assembly of DNA-functionalized nanoparticles. A hybridized double stranded (ds) DNA is formed when complementary nucleobases on hybridizing strands exhibit specific and directional hydrogen bonds through canonical Watson-Crick base-pairing interactions. In recent years, the need for cheaper alternatives and significant synthetic advances have driven design of DNA mimics with new backbone chemistries. However, a fundamental understanding of how these backbone modifications in the oligo-nucleic acids impact the hybridization and melting behavior of the duplex is still lacking. In this talk, we present our recent findings on impact of varying backbone chemistry on hybridization of oligo-nucleic acid duplexes. We use coarse-grained molecular dynamics simulations to isolate the effect of strand flexibility, electrostatic interactions and nucleobase spacing on the melting curves for duplexes with various strand sequences and concentrations. Since conjugation of oligo-nucleic acids with polymers serve as building blocks for thermo-responsive polymer networks and gels, we also present the effect of such conjugation on hybridization thermodynamics and polymer conformation.   

Pore Diameter Dependence and Segmental Dynamics of Poly-Z-L-lysine and Poly-L-alanine Confined in 1D Nanocylindrical Geometry

Structure formation, thermodynamic stability, phase and dynamic behaviors of polypeptides are strongly affected by confinement. Since understanding the changes in these behaviors will allow their rational design as functional devices with tunable properties, herein we investigated Poly-Z-L-lysine (PZLL) and Poly-L-alanine (PAla) homopolypeptides confined in nanoporous alumina containing aligned cylindrical nanopores as a function of pore size by differential scanning calorimetry (DSC), Fourier Transform Infrared Spectroscopy, Solid-state NMR, X-ray diffraction, Dielectric spectroscopy(DS). Bulk PZLL exhibits a glass transition temperature (T$_{\mathrm{g}})$ at about 301K while PZLL nanorods showed slightly lower T$_{\mathrm{g\thinspace }}$(294K). The dynamic investigation by DS also revealed a decrease (4K) in T$_{\mathrm{g}}$ between bulk and PZLL nanorods. DS is a very sensitive probe of the local and global secondary structure relaxation through the large dipole to study effect of confinement. The results revealed that the local segmental dynamics, associated with broken hydrogen bonds, and segmental dynamics speed-up on confinement.   

Effects of spermine binding on Taxol-stabilized microtubules

Previous studies have shown that polyamines such as spermine present in cells at physiological concentrations can facilitate the polymerization of tubulins into microtubules (MTs). A recent experiment demonstrates that in the presence of high-concentration spermine, Taxol-stabilized MTs undergo a shape transformation into inverted tubulin tubules (ITTs), the outside surface of which corresponds to the inside surface of a regular MT. However, the molecular mechanism underlying the shape transformation of MTs into ITTs is unclear. We perform all atom molecular dynamics simulations on Taxol-stabilized MT sheets containing two protofilaments surrounded by spermine ions. The spermine concentration is varied from 0 to 25mM to match the range probed experimentally. We identify important spermine binding regions on the MT surface and the influence of the spermine binding on the structure and dynamics of MTs. In contrast to Taxol, our results show that spermine binding seems to decrease the flexibility of tubulin proteins, resulting in weaker tubulin-tubulin contacts and promoting the bending of protofilaments into curved protofilaments, inverted rings, and eventually inverted tubules.   

Tuning the entropic spring to dictate order and functionality in polymer conjugated peptide biomaterials

Hybrid peptide-polymer conjugates have the potential to combine the advantages of natural proteins and synthetic polymers, resulting in biomaterials with improved stability, controlled assembly, and tailored functionalities. However, the effect of polymer conjugation on peptide structural organization and functionality, along with the behavior of polymers at the interface with biomolecules remain to be fully understood. This talk will summarize our recent efforts towards establishing a modeling framework to design entropic forces in helix-polymer conjugates and polymer-conjugated peptide nanotubes to achieve hierarchical self-assembling systems with predictable order. The first part of the talk will discuss how self-assembly principles found in biology, combined with polymer physics concepts can be used to create artificial membranes that mimic certain features of ion channels. Thermodynamics and kinetics aspects of self-assembly and how it governs the growth and stacking sequences of peptide nanotubes will be discussed, along with its implications for nanoscale transport. The second part of the talk will review advances related to modeling polymer conjugated coiled coils at relevant length and time scales. Atomistic simulations combined with sampling techniques will be presented to discuss the energy landscapes governing coiled-coil stability, revealing cascades of events governing disassembly. This will be followed by a discussion of mechanisms through which polymers can stabilize small proteins, such as shielding of solvents, and how specific peptide sequences can reciprocate by altering polymer conformations. Correlations between mechanical and thermal stability of peptides will be discussed. Finally, coarse-grained simulations will provide insight into how the location of polymer attachment changes entropic forces and higher-level organization in helix bundle assemblies. Our findings set the stage for a materials-by-design capability towards dictating complex topologies of polymer-peptide conjugate systems.   

Thermal Properties of Silk Fibroin Using Fast Scanning Calorimetry

We performed fast scanning chip-based calorimetry of silk protein using the Mettler Flash DSC1. We suggest the methodology by which to obtain quantitative information on the very first scan to high temperature, including the melting endotherm of the beta pleated sheets.~For proteins, this first scan is the most important one, because the crystalline secondary structural features, the beta pleated sheets, melt after the first heating and cannot be thermally reintroduced.~To obtain high quality data, the samples must be treated to drying and enthalpy relaxation sequences.~ The heat flow rates in heating and cooling must be corrected for asymmetric heat loses.~We evaluate methods to obtain an estimate of the sample mass, finally choosing internal calibration using the known heat capacity increment at the glass transition.~We report that even heating at rates of 2000 K/s, thermal degradation of silk cannot be totally avoided, though it can be minimized.~Using a set of nineteen samples, we successfully determine the liquid state heat capacity of silk as: $C_{p}^{liquid}(T )=$ (1.98 $+$0.06) J/gK $+ \quad T$ (6.82 $+$1.4) x10$^{-4}$ J/gK$^{2}$. Methods for estimation of the sample mass will be presented and compared.   

Long-term Controlled Drug Release from bi-component Electrospun Fibers

Multi-drug delivery systems with timed programmed release are hard to be produced due to the complex drug release kinetics which mainly refers to the diffusion of drug molecules from the fiber and the degradation of the carrier. This study focused on the whole life-time story of the long-term drug releasing fibrous systems. Electrospun membrane utilizing FDA approved polymers and broad-spectrum antibiotics showed specific drug release profiles which could be divided into three stages based on the profile slope. With throughout morphology observation, cumulative release amount and releasing duration, releasing kinetics and critical factors were fully discussed during three stages. Through changing the second component, approximately linear drug release profile and a drug release duration about 13 days was prepared, which is perfect for preventing post-operative infection. The addition of this semi-crystalline polymer in turn influenced the fiber swelling and created drug diffusion channels. In conclusion, through adjusting and optimization of the blending component, initial burst release, delayed release for certain duration, and especially the sustained release profile could all be controlled, as well as specific anti-bacterial behavior could be obtained.   

Correlation Between Chain Architecture and Hydration Water Structure in Polysaccharides

The physical properties of confined water can differ dramatically from those of bulk water. Hydration water associated with polysaccharides provides a particularly important example of confined water, with differences in polysaccharide structure providing different spatially confined environments for water adsorption. We have used attenuated total reflection infrared (ATR-IR) spectroscopy to investigate the structure of hydration water in films of three different polysaccharides under controlled relative humidity (RH) conditions. We compare the results obtained for films of highly branched, monodisperse phytoglycogen nanoparticles to those obtained for two unbranched polysaccharides, hyaluronic acid (HA) and chitosan. We find similarities between water structuring in the two linear polysaccharides, and significant differences for phytoglycogen. In particular, the phytoglycogen nanoparticles exhibited high network water connectivity, and a large increase in the fraction of multimer water clusters with increasing RH, whereas the water structure for HA and chitosan was found to be insensitive to changes in RH. These measurements provide unique insight into the relationship between the chain architecture and hydration of polysaccharides.