Homeostasis and spreading depolarization in multiscale simulation of ischemic stroke.

Ionic homeostasis in neurons is essential for maintaining stable electrophysiological properties and a disruption of homeostasis is a characteristic of several pathologies including ischemia and epilepsy. Ionic homeostasis also provides an additional constraint on model parameters. Modeling and simulation of homeostasis with few ions or single compartment neurons can be analyzed mathematically, however for detailed neuronal morphology with multiple charge carrier a numerical approach is required.

Recent developments and performance improvements in the NEURON (neuron.yale.edu) reaction-diffusion module (rxd) allow us to model multiple relevant concentration in the intracellular and extracellular space. We modeled a morphologically detailed hippocampal pyramidal neuron and combine rxd with evolutionary algorithms (BluePyOpt) to obtain parameters that produce realistic electrophysiological responses while maintaining homeostasis of K+, Na+, Ca2+, Cl- and glutamate. We use this model to simulate spreading depression following ischemic stroke.

We developed multiscale models of spreading depolarization in ischemic stroke, coupling electrophysiology and intracellular molecular alterations and bulk tissue alterations mediated by tortuous extracellular diffusion. We simulated spreading depolarization by placing biophysically detailed models of pyramidal neurons in the penumbrae, with reduced pump conductances. We used this model to explore the hypothesis that individual neurons show a pattern of susceptibility to damage due to morphology and physiology, such as the surface area to volume ratio or the intracellular calcium dynamics. In morphologically simplified cells, excessive intracellular Ca2+ produced greater susceptibility in proximal compared to distal dendrites.