OCNS 2026 Workshop

Cerebral ischemia impairs neuronal and glial function, ranging from transient synaptic failure to irreversible damage. The effects of ischemia on excitatory synaptic transmission remain incompletely understood. Here, we present a detailed biophysical model including the first full implementation of the glutamate-glutamine cycle (GG-cycle), which is essential for proper functioning of glutamatergic synapses. 

We simulate a presynaptic neuron and an astrocyte in a finite extracellular space (ECS), surrounded by an oxygen bath, as a proxy for energy supply. The model includes ionic currents with corresponding channels and transporters such as the sodium-potassium ATPase. To model synaptic transmission, we combine calcium-dependent glutamate release, its uptake by the sodium-dependent excitatory amino acid transporters (EAATs), and the GG-cycle, including glutamine synthesis. 

We simulate ischemia by blocking energy supply completely. The neuron enters a depolarization block, ion concentrations reach pathological values, and glutamate accumulates in the ECS while glutamate release is disrupted. Surprisingly, we found that synaptic transmission failure was not primarily caused by excessive glutamate release nor by failure of glutamine synthetase. Instead, it mainly resulted from EAAT dysfunction, driven by the collapse of the sodium gradient. Enhancing glutamate clearance alone was insufficient for recovery of synaptic transmission. However, inhibition of the voltage-gated Na+ channels restored ion gradients, recovered glutamate uptake, and re-enabled glutamate release. Taken together, our study highlights the critical role of ion homeostasis, in particular the sodium gradient, in maintaining synaptic function during metabolic stress. Moreover, the model provides a better understanding of synaptic transmission failure and potential recovery strategies during ischemia.

Cerebral ischemia impairs neuronal and glial function, ranging from transient synaptic failure to irreversible damage. The effects of ischemia on excitatory synaptic transmission remain incompletely understood. Here, we present a detailed biophysical model including the first full implementation of the glutamate-glutamine cycle (GG-cycle), which is essential for proper functioning of glutamatergic synapses. 

We simulate a presynaptic neuron and an astrocyte in a finite extracellular space (ECS), surrounded by an oxygen bath, as a proxy for energy supply. The model includes ionic currents with corresponding channels and transporters such as the sodium-potassium ATPase. To model synaptic transmission, we combine calcium-dependent glutamate release, its uptake by the sodium-dependent excitatory amino acid transporters (EAATs), and the GG-cycle, including glutamine synthesis. 

We simulate ischemia by blocking energy supply completely. The neuron enters a depolarization block, ion concentrations reach pathological values, and glutamate accumulates in the ECS while glutamate release is disrupted. Surprisingly, we found that synaptic transmission failure was not primarily caused by excessive glutamate release nor by failure of glutamine synthetase. Instead, it mainly resulted from EAAT dysfunction, driven by the collapse of the sodium gradient. Enhancing glutamate clearance alone was insufficient for recovery of synaptic transmission. However, inhibition of the voltage-gated Na+ channels restored ion gradients, recovered glutamate uptake, and re-enabled glutamate release. Taken together, our study highlights the critical role of ion homeostasis, in particular the sodium gradient, in maintaining synaptic function during metabolic stress. Moreover, the model provides a better understanding of synaptic transmission failure and potential recovery strategies during ischemia.

Epileptic seizures emerge from complex interactions between neuronal excitability, ionic homeostasis, and cellular metabolism. Although the Na+-K+-ATPase pump is known to play a central role in restoring ionic gradients during neuronal activity, the influence of ATP availability on seizure dynamics remains poorly understood. Here, we extend the Epileptor-2 model by introducing an energy-dependent formulation of Na+-K+-ATPase pump activity coupled to intracellular ATP dynamics. ATP production and consumption are modeled through phenomenological kinetic equations linking metabolic state to ionic regulation.

Using experimentally inspired ATP-related fluorescence signals, we calibrate the ATP production parameters to reproduce realistic depletion and recovery timescales during seizure-like events (SLEs). Within this framework, we show that ATP production acts as a critical control parameter governing transitions between distinct pathological regimes. Reduced ATP production capacity prolongs seizures and induces transitions from periodic SLEs to status epilepticus-like events (SELEs), characterized by sustained elevations of extracellular potassium and persistent neuronal firing. Severe metabolic impairment further traps the system in depolarized pathological states.

We additionally investigate the influence of astrocytic potassium buffering through bifurcation analysis of the glial uptake parameter $G_{glia}$. The analysis reveals transitions between stable resting states, periodic seizure activity, SELEs, and spreading depression-like events (SDLEs). In SDLE regimes, ATP transiently decreases during the onset of depolarization before recovering as neuronal firing collapses and pump activity diminishes.

The neural mechanisms underlying the emergence of functional connectivity in resting-state fMRI remain poorly understood. Recent studies suggest that resting-state activity consists of brief periods of strong co-fluctuations among brain regions, which reflect overall functional connectivity. Others report a continuum in co-fluctuations over time, with varying degree of correlation to functional connectivity. These findings raise the critical question: what neural processes underlie the temporal structure of resting-state activity? To address this, we used a biophysically realistic whole-brain computational model in which resting-state activity emerged from temporal variations in the ion concentrations of potassium (K+) and sodium (Na+), intracellular chloride (Cl-), and the activity of the Na+/K+ ATPase. The model reproduced transient periods of high co-fluctuations, and the functional connectivity at different co-fluctuation levels correlated to varying degrees with the connectivity measured over the entire simulation, in line with experimental observations. The periods of high co-fluctuations were aligned with large changes in extracellular ion concentrations. Furthermore, critical parameters governing ion dynamics strongly affected both the timing of these transient events and the spatial structure of the resulting functional connectivity. The balance of excitatory and inhibitory activity further modulated their frequency and amplitude. Together, these results suggest that intrinsic fluctuations in ion dynamics could serve as a plausible neural mechanism for the temporal organization of co-fluctuations and resting-state functional connectivity.

This presentation investigates the neuromodulatory influence of norepinephrine on the biophysical properties of the thalamocortical network, specifically addressing how noradrenergic signalling disrupts the rhythmic oscillations characteristic of Non-Rapid Eye Movement (NREM) sleep, such as slow-wave activity and sleep spindles. The Locus Coeruleus (LC), a brainstem nucleus, serves as the primary efferent source of norepinephrine, providing diffuse projections across the neocortex to activate alpha- and beta-adrenergic receptors. This activation initiates intracellular enzymatic pathways that primarily modulate Hyperpolarization-activated cyclic nucleotide-gated (HCN) and potassium leak channels, shifting the dynamics of potassium, but also calcium-mediated currents. Consequently, this modulation induces sustained neuronal depolarization and the subsequent suppression of spindle activity. This study extends the current state-of-the-art spiking neural network model of sleep by incorporating the neuromodulatory effects of norepinephrine at the cellular level [1]. Transient fluctuations in NA concentrations are simulated to reproduce tonic and phasic Locus Coeruleus (LC) activity on network dynamics. The simulations results are compared against simultaneous in vivo electrocorticography (ECoG) and multi-unit LC recordings obtained from rats during natural sleep cycles. 

[1] Fink CG, Sanda P, Bayer L, Abeysinghe E, Bazhenov M, Krishnan GP. Python/NEURON code for simulating biophysically realistic thalamocortical dynamics during sleep. Softw Impacts. 2024 Sep;21:100667. doi: 10.1016/j.simpa.2024.100667