Kinetic modeling of amyloid fibrillation and synaptic plasticity as memory loss and formation mechanisms

Abstract

The principles of biochemical kinetics and system engineering are applied to explain memory-related neuroscientific phenomena. Amyloid fibrillation and synaptic plasticity have been our focus of research due to their significance. The former is related to the pathology of many neurodegenerative diseases and the later is regarded as the principal mechanism underlying learning and memory. Claimed to be the number one cause of senile dementia, Alzheimer's disease (AD) is one of the disorders that involve misfolding of amyloid protein and formation of insoluble fibrils. Although a variety of time dependent fibrillation data in vitro are available, few mechanistic models have been developed. To bridge this gap we used chemical engineering concepts from polymer dynamics, particle mechanics and population balance models to develop a mathematical formulation of amyloid growth dynamics. A three-stage mechanism consisting of natural protein misfolding, nucleation, and fibril elongation phases was proposed to capture the features of homogeneous fibrillation responses. While our cooperative laboratory provided us with experimental findings, we guided them with experimental design based on modeling work. It was through the iterative process that the size of fibril nuclei and concentration profiles of soluble proteins were elucidated. The study also reveals further experiments for diagnosing the evolution of amyloid coagulation and probing desired properties of potential fibrillation inhibitors. Synaptic plasticity at various time ranges has been studied experimentally to elucidate memory formation mechanism. By comparison, the theoretical work is underdeveloped and insufficient to explain some experiments. To resolve the issue, we developed models for short-term, long-term, and spike timing dependent synaptic plasticity, respectively.

(cont.) First, presynaptic vesicle trafficking that leads to the release of glutamate as neurotransmitter was taken into account to explain short-term plasticity data. Second, long-term plasticity data lasting for hours after tetanus stimuli has been matched by a calcium entrapment model we developed. Model differentiation was done to demonstrate the better performance of calcium entrapment model than an alternative bistable theory in fitting graded long-term potentiation responses. Finally, to decipher spike timing dependent plasticity (STDP), we developed a systematic model incorporating back propagation of action potential, dual requirement of NMDA receptors, and calcium dependent plasticity. This built model is supported by five different types of STDP experimental data. The accumulation of amyloid beta has been found to disrupt the sustainable modification of long-term synaptic plasticity which might explain the inability of AD patients to form new memory at early stage of the disease. Yet the linkage between the existence of amyloid beta species and failure of long-term plasticity was unclear. We suggest that the abnormality of calcium entrapment function caused by amyloid oligomers is the intermediate step that eventually leads to memory loss. Unsustainable calcium level and decreased postsynaptic activities result into the removal or internalization of alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors. The number of AMPA receptors as the indicators of synaptic strength may result into disconnection between neurons and even neuronal apoptosis. New experiments have been suggested to validate this hypothesis and to elucidate the pathology of Alzheimer's disease.