Predicting excitotoxicity at the onset of Alzheimer’s disease by a model of Aβ-dependent trafficking of astrocytic glutamate transporters

At Alzheimer’s disease (AD) onset, extracellular accumulation of oligomeric amyloid-β (Aβ) correlates with excitotoxicity and the alteration of glutamate uptake by astrocytic transporters (GLT1). Experiments suggest that glutamate, Aβ, or a combination thereof, may dynamically regulate trafficking and expression of those transporters between perisynaptic and intracellular astrocytic compartments [1]. There is no understanding however, whether and how such mechanisms could ultimately link with the emergence of excitotoxicity which hallmarks early stages of AD. With this regard, we consider a simplified description of astrocytic transporter trafficking based on a Markov process for transporter movements between the cytoplasm and the plasma membrane, and vice versa, and we use this model to identify potential ensembles of Aβ-dependent pathways of trafficking that could account for experimental observations. Next, we consider the mean-field rate description ensuing from the Markov model in the Finite Element Method (FEM) framework of a 3D model of glutamate diffusion at synaptic terminals and their surroundings. Changing extracellular Aβ concentration, we accordingly look at the time course of extracellular glutamate, and the conditions for its accumulation in the extrasynaptic space. Since extracellular glutamate alters synaptic activity, we estimate conditions for excitotoxicity linking neural network firing activity with extracellular glutamate. Consistent with experiments, our model predicts that GLT1 surface expression decreases when extracellular Aβ increases beyond a threshold concentration. This, in turn, favors extracellular glutamate accumulation, promoting a positive feedback loop that induce further synaptic glutamate release, and thereby excitotoxicity. Because the rate of glutamate accumulation depends on the uptake capacity by astrocytic transporters, which is a function of extracellular Aβ, local gradients of Aβ may dramatically affect synaptic environment, both locally and extra-synaptically. These results provide theoretical support to the possibility of looking at Aβ-dependent astrocytic GLT1 expression as a clinical marker for early diagnosis Alzheimer’s disease.