Computational insights into multivalently binding polymers
Massachusetts Institute of Technology. Department of Materials Science and Engineering.
MetadataShow full item record
Multivalent binding is commonly used throughout biology to create strong, conformal bonds using multiple weak binding interactions simultaneously. Bonds are considered multivalent when multiple ligands on one species simultaneously bind to multiple receptors on another species. Together, this bond can be much stronger than the sum of its parts. Throughout this thesis, we use theory and coarse-grain Brownian dynamics simulations with specific reactive-binding to explore general characteristics of multivalent polymer interactions. Our simulations bridge length and timescales and can sample large polymer systems that bind targets at the sub-nanometer lengthscale. While the simulation and theory presented is very general and can be applied to many different systems of multivalent polymers, this thesis specifically explores consequences for two applications: multivalent polymers as decoys to inhibit infection and polymers as scaffolds for biocondensates.Many pathogens use multivalent bonds to attach to our cell surfaces before entering and causing infection. Therefore, there is significant interest in preventing infection from viruses, bacteria, and toxic proteins by inhibiting this attachment step using multivalent decoys. There have been many experiments showing successful binding of long polymers or other large multivalent architectures to colloids or small proteins that pathogens use to bind to our cells. While these experiments have shown how promising multivalent inhibitors are for preventing infection, a theoretical understanding of why design parameters of multivalent polymers result in a particular binding affinity is still missing. Simulations can easily isolate a single design parameter to provide direct links between structure and function, when experiments cannot always do so. This research is intended to provide a systematic study linking structure of multivalent polymers to their binding behavior.In the first half of this thesis, we explore design properties of polymeric binders and how degree of polymerization, solvent quality, binding site affinity patterns, backbone stiffness, and target concentration change the multivalent binding affinity. We provide simple theory to show that multivalent polymers are limited by their ability and the energetic costs of forming polymer loops. We go on to show how these results and theory have implications on the binding affinity of polymers with heterogeneous binding sites and determines the effect of polymer backbone flexibility and solvent quality on binding affinity. Multivalent polymers are also an essential component of biocondensates, liquid-like droplets comprised of proteins and nucleic acids are found throughout cells. Although the function of these biocondensates is still an active field of study, it is clear that multivalent polymers are essential to their formation through liquid-liquid phase separation (LLPS).There is little theoretical study of biocondensates that contain binding between species of asymmetric size and valency and the effects of multivalent polymers on the dynamics of these liquid droplets is not well understood. Studying how multivalent polymers modulate droplet dynamics is important because droplet crystallization or solidification is often associated with neurodegenerative disorders such as dementia and amyotrophic lateral sclerosis (ALS). Therefore, in the second half of this thesis, we present research on the role multivalent polymers play in LLPS droplets and their resulting dynamics. We consider how a host of design parameters can change the phase boundary of systems with multivalent polymers binding to smaller targets including solvent quality, valency, binding affinity, specific versus non-specific binding sites, and backbone stiffness.We found that consistent with previous work on other systems, asymmetric valency systems also showed increased phase separation with increased binding affinities and valencies. We show that phase separation due to non-specific bonds is highly sensitive to changes in attraction, but that phase separation through specific-bonds is much more robust. By combining specific and non-specific multivalency, systems can precisely tune the phase separation boundary. Polymer stiffness can also modulate the phase boundary, where stiff, rod-like polymers were less able to cause phase separation than their flexible counterparts. We also elucidate how polymer-target binding affinities can be used to form micro-phase separated droplets. Lastly, we show that increasing attraction to polymers can slow target diffusion inside droplets while decreasing the density of droplets, with implications for droplet solidification.We hope that this work will provide direction for the rational design of synthetic multivalent polymer systems such as pathogen inhibitors as well as improve understanding of native biological systems like biocondensates.
Thesis: Ph. D., Massachusetts Institute of Technology, Department of Materials Science and Engineering, 2020Cataloged from student-submitted PDF of thesis.Includes bibliographical references (pages 253-264).
DepartmentMassachusetts Institute of Technology. Department of Materials Science and Engineering
Massachusetts Institute of Technology
Materials Science and Engineering.