Wolynes, Peter G2022-12-212022-12-212022-122022-12-02December 2Gu, Xinyu. "Multi-scale computational modeling of an RNA-binding prion, CPEB3, reveals its molecular mechanisms underlying the formation of long-term memory." (2022) Diss., Rice University. <a href="https://hdl.handle.net/1911/114183">https://hdl.handle.net/1911/114183</a>.https://hdl.handle.net/1911/114183The growth and stabilization of dendritic spines is thought to be essential for strengthening the connections between neurons, and thereby memories. Actin cytoskeleton remodeling in spines is the basis of this growth and stabilization. CPEB proteins were first identified as a group of RNA-binding proteins regulating the translation of their target mRNAs, like actin mRNAs. Intriguingly, one isoform in CPEB family, CPEB3, has been recently reported as a functional prion that interacts with actin cytoskeleton. These observations make CPEB3 a seductively plausible candidate as a synaptic tag to strengthen the actin cytoskeleton and long-term memory by forming stable aggregates, and simultaneously regulate the local translation of synaptic proteins in spines. Numerous gene knockout experiments have been conducted about CPEB3 and its homologs to investigate CPEB3's function in memories. However, it is challenging for experimentalists to collect structural information of CPEB3 due to the conformational flexibility of prion-like proteins, and the picture of CPEB3's role in synaptic plasticity is still unclear at molecular level. In this thesis, to fill in the missing pieces in the molecular mechanisms of CPEB3, we utilized multi-scale computational modeling by conducting bioinformatic searches, setting up reaction-diffusion systems, and mainly by running molecular dynamics simulations using a coarse-grained protein force field - the Associative memory, Water-mediated, Structure and Energy Model (AWSEM). In the first part, we studied the interaction between actin and CPEB3 and proposed a molecular model for the complex structure of CPEB3 bound to an actin filament (F-actin). Our model gives insights into the molecular details of the F-actin/CPEB3 positive feedback loop underlying long-term memory which involves CPEB3's binding to F-actin, its aggregation triggered by F-actin, and its regulation by SUMOylation. The soluble CPEB3 monomers repress translation, whereas CPEB3 aggregates activate the translation of its target mRNAs. The CPEB3 aggregates, however, that act as long-lasting prions providing "conformational memory", may raise the problem of the consequent translational activation being unregulated. In the second part of the thesis, I propose a computational model of the complex structure between CPEB3 RNA-binding domain (CPEB3-RBD) and small ubiquitin-like modifier protein 2 (SUMO2). Free energy calculations suggest that the allosteric binding of CPEB3 with SUMO2 can confine the CPEB3-RBD to a conformation that favors RNA-binding, and thereby can amplify its RNA-binding affinity. Combining this model with previous experiments showing that CPEB3 monomers are SUMOylated in basal synapses but become deSUMOylated and start to aggregate upon stimulation, we suggest a way in which the translational control of CPEB3 can be switched back to a repressive mode after a stimulation pulse, through an RNA binding shift from binding to CPEB3 fibers to binding to SUMOylated CPEB3 monomers in basal synapses. In the last part, inspired by the specific geometry and polarity of the assembly of mRNAs and CPEB3 aggregates, a vectorial channeling mechanism is proposed to describe the local translational regulation by general mRNA/protein assemblies including functional prions and condensates. The analysis shows that the vectorial processive nature of translation can couple to transport via diffusion so as to repress or activate translation depending on the structure of the RNA protein assembly. We find that multiple factors including diffusivity changes and free energy biases in the assemblies can regulate the translation rate of mRNA by changing the balance between substrate recycling and competition between mRNAs.application/pdfengCopyright is held by the author, unless otherwise indicated. Permission to reuse, publish, or reproduce the work beyond the bounds of fair use or other exemptions to copyright law must be obtained from the copyright holder.Functional prionDendritic spinesBiomolecular condensateThesis2022-12-21