Browsing by Author "Arya, Gaurav"

Many vital biological tasks, such as protein degradation, DNA strand separation, and viral DNA packaging are performed by ring NTPase assemblies. These assemblies harvest energy from NTP binding and hydrolysis in order to translocate their biopolymer substrate through their central pores. Single-molecule characterization demonstrated that these assemblies are highly coordinated and produce forces an order of magnitude larger than most molecular motors. Recently, many structures of these assemblies have been experimentally solved and resulting globular translocation models have been proposed. While these static structures have provided great insights into how molecular motors assemble, the specific molecular mechanisms that promote, regulate, and coordinate the dynamic translocation processes remain poorly understood. In this dissertation, I use computational tools to model ring ATPase molecular motors in order to elucidate such mechanisms. Initially, I focus on viral packaging ATPases and then generalize my findings to a broader class of motors by studying FtsK-like and AAA+ motors. For all systems, atomistic molecular dynamics simulations were used to calculate free-energy landscapes that predict conformational changes, predict mutual-information-based signaling pathways that couple enzymatic and mechanical activities, predict principal components of motion that describe the enzyme’s native function, and predict the effects of mutagenesis in silico. For viral packaging ATPases, I first predicted that a strictly conserved Walker A arginine residue functions analogously to a sensor II motif arginine found in AAA+ systems, and that it is used to couple ATP binding to lid subdomain rotation. Second, I predicted how mutations in the Walker A and Walker B motifs could abrogate enzyme function. All these predictions were corroborated by collaborators’ extensive experimental characterization. Third, I helped build the first structure of an actively packaging viral ATPase motor into the cryo-EM reconstruction and led the biological interpretation of the resulting structure. Fourth, I used molecular dynamics simulations of pentameric ATPase assemblies to predict how the assemblies respond to nucleotide-occupancy and presence of double-stranded DNA substrate. Based on the structure and simulations, I proposed the helical-to-planar model of viral DNA packaging, which is the first atomistic model that can predict the salient features of viral DNA packaging. Further, this model lays the groundwork of future work by predicting specific conformational changes and interactions that were otherwise obscure from experimental studies. Fifth, I tested a key proposal in my helical-to-planar model by using molecular dynamics simulations to investigate how nucleotide binding is coupled to substrate gripping. The resulting glutamate switch signaling pathway was corroborated by structural data and functional mutagenesis assays. Lastly, I investigated FtsK-like and AAA+ enzymes to probe for molecular mechanisms common to a broad class of translocating ring ATPases. From these studies, I identified a core set of principles that can be modularly added together to describe a number of different translocation models. In summary, the results presented in this dissertation describe fundamental mechanisms of translocating ring ATPase motors. When possible, my computational predictions were corroborated by experimental characterization. When experimental characterization was not yet possible, my predictions and derived models serve as a guide for future studies. The models I derived provide the first comprehensive description of the coordinated conformational changes that drive viral DNA packaging. Further, they have the potential to inform rational design of synthetic molecular motors and anti-viral therapeutics that target the genome packaging step.

Polymer nanocomposites have attracted considerable scientific and technological interest, as such composites combine desirable material properties of both the polymer and the nanoparticles (NPs). New applications of composites often require higher-order, low-dimensional (anisotropic) organization of NPs in polymers, e.g., 1d strings, percolating networks, or 2d sheets. While self-assembly provides a powerful bottom-up approach for fabricating higher-order nanostructures, achieving unique low-dimensional assemblies of NPs in polymers is challenging since NPs tend to self-assemble into three-dimensional close-packed aggregates to minimize their total free energy. In this dissertation, I tackle this challenge of achieving anisotropic NP assembly in polymers through molecular dynamics (MD) simulations along with global optimization and machine learning techniques. First, I present a new strategy for assembling NPs into anisotropic architectures in polymer matrices, which takes advantage of the interfacial tension between two mutually immiscible polymers forming a bilayer and differences in the relative miscibility of polymer grafts with the two polymer layers to trap NPs within 2d planes parallel to the interface. Coarse-grained MD simulations are used to demonstrate this strategy, where I illustrate the assembly of NP clusters, such as trimers with tunable bending angle and anisotropic macroscopic phases, including serpentine and branched structures, ridged hexagonal monolayers, and square-ordered bilayers. The above MD simulations are however inefficient for determining the equilibrium structures of NP assemblies, especially those with many particles or complex unit cells. I adapt the efficient Basin-hopping Monte Carlo algorithm to locate the global minimum-energy configurations of NPs at interface, which allows us to explore the full breadth of NP structures possible at interface and discover many unique NP, such as binary superlattices, several of which are yet to be experimentally realized. While exploring the assembly of polymer-grafted NPs at polymer interfaces using explicit coarse-grained MD simulations, we observe that multi-body effects play an important role in the formation of quasi-1d structures. Motivated by this observation, and by similar observations in bulk polymer, I introduce a general machine learning (ML) approach to develop an analytical potential that can describe many-body interactions between polymer-grafted NPs in a polymer matrix, where the high-dimensional energy landscape of NPs is fitted by permutationally invariant polynomials as a function of their interparticle distances. The developed potential reduces the computational cost by several orders of magnitude and thus allows us to explore NP assembly at large length and time scales. Lastly, I investigate the orientational behaviors of shaped NPs (cubic NPs) at interfaces. I demonstrate the possibility of tuning the orientations of nanocubes between all three orientation phases (face up, edge up and vertex up) through polymer grafts and then take advantage of their orientational effects to assemble them into unique clusters, such as rectilinear strings, close-packed sheets, bilayer ribbons, and perforated sheets. Furthermore, by using two species of grafts, where one is hydrophilic and the other is hydrophobic, I demonstrate that the interactions between nanocubes can be further manipulated by controlling the length and stoichiometry of the two grafts, leading to more open, reconfigurable NP assemblies. Overall, this dissertation suggests that interfacial assembly of NPs could be a promising approach for fabricating next-generation functional materials with potential applications in plasmonics, electronics, optics, and catalysis.

DNA nanotechnology is a fascinating field that eschews using DNA as an information storage medium and instead uses it as a nanoscale structural material, taking advantage of the canonical base pairing rules to fold DNA into shapes, patterns, and mechanical devices 10,000 times smaller than a human hair. Over the 40 years of the field's existence, DNA nanotechnology has progressed from building simple wireframe structures to a full-blown nanoengineering ecosystem with the ability to construct logic-gated nanoscale drug delivery vehicles, computing devices, robots, and more. Key to the development of the field has been the growing ability to predict the behavior of DNA nanostructures. However, much is still not understood these devices' self-assembly and dynamic behaviors. The reason for this is that DNA interacts on a short length scale and a long timescale, and many processes occur far from equilibrium, both of which make modeling their behavior challenging. This dissertation presents three projects employing mesoscopic simulations, statistical mechanics, and numerical free energy landscape calculations to provide access to these length and time scales in order to better understand the self-assembly and organization of DNA nanostructures. Specifically, mesoscopic simulations are used to directly simulate the self-assembly of DNA nanostructures and understand the mechanism of their folding; lattice simulations are used to understand the phase behavior of arrays of molecular rotors made from DNA; and geometric calculations and Brownian dynamics simulations are used to computationally derive a bottom-up technique for templating heterogeneous DNA origami species on a single lithographically-defined template.

While anisotropic nanoparticles provide unique building blocks for self-assembling useful nanodevices and nanomaterials ranging from plasmonic sensors to chiral metamaterials, controlling their self-assembly process to achieve targeted structure remains challenging. Recently, surface functionalization of nanoparticles with polymer grafts was shown to be a powerful strategy for tuning the orientation-dependent interactions of the nanoparticles. This technique allows modulation of the interaction between nanoparticles as grafted polymers can provide both repulsive interactions arising from their steric hindrance as well as attractive interactions due to their adsorption to the particle surfaces. Utilizing this approach, experiments have successfully assembled nanoparticles into large structures with highly uniform interparticle orientations. However, many challenges remain in fabricating desired nanostructures with the polymer-grafted anisotropic nanoparticles. First, much of the underlying physics governing assembly of such nanoparticles is not well understood and is difficult to discern using experimental techniques due to the nanoscopic nature of the self-assembly process. Second, the relevant parameter space that affects the particle assembly is vast and investigation of such large parameter space is costly in terms of both time and expenses. Third, computationally investigating the behavior of anisotropic nanoparticles is difficult as calculation of their interaction energies is computationally expensive due to the lack of analytical expressions for these energies.In this dissertation, I tackle these challenges in self-assembly of anisotropic nanoparticles through computational modeling, focusing specifically on polymer-grafted nanocubes and DNA-grafted nanorods. For both systems, computational methods and analytical models for efficiently calculating the interaction energies between the anisotropic nanoparticles are first developed. Using such methods as well as advanced Monte Carlo simulations and atomistic calculations, free-energy landscapes describing the assembly of these anisotropic nanoparticles are obtained. Analysis of the free-energy landscapes demonstrates that understanding the interplay between the different interaction components of the systems as well as their dependencies on the relative configurations of the assembled particles is crucial. Specifically for the nanocubes, the competition between the attractive interactions between the inorganic particle cores lead to face-face type of configurations while the repulsive interactions due to the polymer corona induce edge-edge configurations. For the DNA-grafted nanorods, the competition between attractive and repulsive interactions interplay with the chirality of the bridging DNA to induce chiral assembly of the nanorods. Based on these results, material design rules for assembling both the nanocubes and the nanorods into desired configurations are suggested. These results were not only in agreement with many previous experimental studies but also provided the underlying mechanism that explain such assembly behaviors. In summary, the results presented in this dissertation should both aid in fabrication of nanodevices with precisely controlled particle assemblies as well as provide efficient computational methods for future investigation of anisotropic nanoparticles.